ORNL/EIS-189
EPA-560/11-80-028
Contract No. W-7405-eng-26
PROCEEDINGS OF THE WORKSHOP ON
SUBCHRONIC TOXICITY TESTING
Denver, Colorado
May 20-24, 1979
Edited by
Norbert Page, D.V.M.
Daljit Sawhney, D.V.M., Ph.D.
Office of Pesticides and Toxic Substances
U.S. Environmental Protection Agency
Washington, D.C. 20460
and
Michael G. Ryon
Health and Environmental Studies Program
Information Center Complex/Information Division
Oak Ridge National Laboratory
Oak Ridge, Tennessee 37830
Work sponsored by the Office of Pesticides and Toxic Substances,
U.S. Environmental Protection Agency, Washington, D.C., under
Interagency Agreement No. 80-D-X0453
Date Published: November 1980
OAK RIDGE NATIONAL LABORATORY
OAK RIDGE, TENNESSEE 37830
operated by
UNION CARBIDE CORPORATION
for the
DEPARTMENT OF ENERGY
-------
-------
At the time of publication, we were saddened to learn of the death of a workshop
contributor, Bernard McNamara. His scientific and personal advice will be missed by
those who worked with him on this and other projects.
in
-------
-------
CONTENTS
Tables vii
Preface ix
Keynote Address to the Workshop Participants 1
Warren R. Muir
Review of Current Activities in Test Standards Development 5
Norbert P. Page
Charge to the Workshop and the Individual Committees 11
Wayne M. Calbraith
Proceedings of the Workshop on Subchronic Toxicity Testing 15
1. Introduction 15
1.1 Scope 15
1.2 General Principles 15
1.3 Objectives 16
2. Relationship of Protocol Design to Chemical Characteristics 16
3. Experimental Design 19
3.1 Exposure and Duration 19
3.2 Dose 20
3.2.1 Selection of Doses 20
3.2.2 Limitations of Maximum Levels 20
3.3 Species Selection 21
3.4 Number of Animals 21
3.5 Age of Animals 21
4. Evaluation of Toxic Effects 22
4.1 Introduction 22
4.2 Clinical Observations and Evaluations 23
4.2.1 Physical 23
4.2.2 Health 23
4.2.3 Cardiovascular 23
4.2.4 Reproductive 24
4.2.5 Ocular 24
4.2.6 Behavioral 24
4.2.7 Neurological 25
4.2.8 Memory and Learning 25
-------
VI
4.3 Clinical Laboratory Tests 25
4.3.1 Sampling Frequency 26
4.3.2 Hematologic Evaluations 26
4.3.3 Clinical Chemistry Evaluation 27
4.4 Special Organ Evaluation 28
4.5 Unnalysis Evaluation 29
4.6 Pathology 29
4.6.1 Gross Examination 29
4.6.2 Tissue Preservation 30
4.6.3 Justification of Selection of Tissues
for Histopathological Examinations 31
4.6.4 Special Considerations 32
4.6.5 Central Nervous System 33
4.7 Analysis and Presentation of Test Results 33
4.8 Usefulness and Correlation of Test Results 33
5. Criteria for Data Extrapolations from One Route of
Administration to Another 35
5.1 Introduction 35
5.2 Review of Data 35
5.3 Factors Affecting Extrapolation of Data Obtained by
One Route of Exposure to Another 40
5.4 Assumptions and Criteria for Extrapolation 41
5.5 Retrospective Analysis and Extrapolation 42
5.6 Prospective Analysis and Extrapolation 43
5.7 Summary and Conclusions 44
6. Limitations of Acute and Subchronic Toxicity Studies 44
6.1 Introduction 44
6.2 Acute Toxicity 45
6.3 Interpretation and Limitations of Acute Toxicity Studies 45
6.4 Subchronic Toxicity 46
6.5 Special Considerations 46
6.5.1 Reproduction and Teratology 46
6.5.2 Recovery 46
6.5.3 Pharmacokinetic and Metabolic Data 47
6.6 Dermal and Ocular Studies 47
6.7 Conclusions 48
7. Unresolved Issues and Research Recommendations 48
7.1 Unresolved Issues 48
7.2 Research Recommendations 49
Appendix: Committees and Participants 53
Committees 53
Participants 55
Bibliography 59
-------
TABLES
1. Experimental design for recently proposed subchronic oral toxicity
tests 8
2. Chemical and physical information 17
3. Acute toxicity information 17
4. General observations, clinical laboratory tests, and pathology
examinations that may be used in subchronic toxicity studies 34
5. Acute toxicity of nefopam in three species 36
6. Effects of administration of compounds in which lethal toxicity is
independent of (isoniazid), partially dependent on
(DFP and pentobarbital), and completely
dependent on (procaine) route of administration 37
7. Comparative results of 90-d toxicity studies in rats administered
vinylidene chloride by drinking water or inhalation 38
8. Comparative toxicity in rats administered perchloroethylene
by gavage or inhalation 38
9. Summary of results from two chronic toxicity studies
in rats with 1,4-dioxane 39
Vll
-------
-------
PREFACE
This document is the product of a three-day workshop sponsored by the Office of
Pesticides and Toxic Substances of the Environmental Protection Agency (EPA) to
evaluate the subchronic toxicity test and its role in chemical assessments mandated
under the Toxic Substances Control Act (TSCA). The Health and Environmental
Studies Program of the Oak Ridge National Laboratory (ORNL) had the responsibility
for the coordination of the workshop and the preparation of the resulting proceedings.
The participants, selected to achieve a balance of industrial, academic, and
governmental viewpoints, were provided with a review of the pertinent literature
prepared by ORNL. Because many of the topics scheduled for discussion had limited
support in available literature, much of the document represents the results of the
committees' discussions, analyses, and recommendations.
The workshop was divided into five committees, each with a specific area for
discussion. The proceedings of the workshop are a consolidation of the five
committee drafts into one report, thereby enhancing the cohesiveness of the rationale
and avoiding repetition in the discussions.
In preparing this report, both during and after the workshop, several individuals
were especially helpful. Dr. Witschi of the ORNL Biology Division advised the ORNL
staff on technical matters and was responsible, in large part, for the initial organization
of the workshop and steering committee. The steering committee members, Drs.
Galbraith, Gibson, Oehme, Peck, Pfitzer, Page, Sawhney, and Terhaar, contributed
much time and effort to the refinement and completion of this project. The success of
the enterprise was also due to the guidance provided by the EPA Deputy Assistant
Administrator, Dr. Muir, and the EPA Senior Medical Advisor, Dr. Seifter. Foremost
among the reasons for the success of the workshop, however, was the enthusiastic
and altruistic attitudes of the participants, who, despite differences in scientific
opinion, maintained friendly, open-minded discussions.
Because the EPA desired a general evaluation of subchronic toxicity testing, this
workshop addressed the major issues. No attempt was made to cover every aspect of
assessment. That some issues are not addressed in the report does not mean they are
less significant. The committees identified areas in which research is needed and
problems of test design unresolved during the workshop. Recommendations based
on these findings were an important product of the committee discussions.
IX
-------
-------
KEYNOTE ADDRESS
TO THE
WORKSHOP PARTICIPANTS
WARREN R. MUIR
U.S. Environmental Protection Agency
I would like to welcome all of you here to Denver and thank you for participating
in this workshop. It is an undertaking that we at the Office of Toxic Substances and
the Environmental Protection Agency (EPA) feel to be very important. We certainly
appreciate that you are attending and helping us with this effort.
I am Warren Muir and I head the Office of Testing and Evaluation, which is one of
the three offices set up to implement the Toxic Substances Control Act (TSCA) in
EPA. As most of you are aware, TSCA was passed in 1976 as a particularly important
new piece of legislation dealing with chemical substances and their hazards. Unlike
the 15 or so preceding acts, which also address regulation of chemical substances,
TSCA is unique in its authorities to generate health, environmental, and other
chemical information from the industry which manufactures or proposes to
manufacture chemicals. Therefore, in our efforts to support our direct regulatory
controls and to increase our understanding of the nature of the chemicals before they
enter our environment, we are especially emphasizing the TSCA aspects of reporting,
record keeping, information systems, testing, and evaluation of information.
The Office of Testing and Evaluation, the scientific office of the three
implementing offices, is responsible for the definition of testing in terms of both
requirements and guidelines. The Office is also responsible for compiling information
available to the EPA (including information to be generated under this testing),
analyzing it, and evaluating it to formulate risk assessments on chemicals.
With respect to testing, there are two provisions that are applicable and in which
we are heavily involved. One is Section 4 of TSCA, a section of the act which allows
EPA to specify both test standards and chemicals to which the test standards would
apply. It basically requires either processors or manufacturers or both to conduct
testing on high priority chemicals. These are chemicals which the EPA may define as
posing an unreasonable risk or which are produced in substantial quantities with
either substantial releases to the environment or substantial risk for human exposure.
With respect to Section 4, the first set of test standards, which include
oncogenicity, chronic toxicity, and the combined oncogenicity and chronic toxicity
protocols, were published in proposed form in the Federal Register about two weeks
ago [9 May 1979]. Also published were the Good Laboratory Practices that were
patterned after the proposed Food and Drug Administration's Good Laboratory
-------
Practices for nonclinical health effects studies. With respect to the topic for this
meeting, subchronic toxicity testing, we intend to propose, in a couple of months,
subchronic toxicity test protocols along with acute toxicity, mutagenicity, reproduc-
tion, and teratology tests. The proposal, which will be published in the Federal
Register, will be a reproposal of the test protocols that have been previously proposed
by the Office of Pesticides Programs under Section 3 of their pesticide guidelines for
the Federal Insecticide, Fungicide, and Rodenticide Act. We are proposing those
standards in an effort to focus upon the various issues associated with those test
protocols and to allow us the time to get together with the Office of Pesticides so that
in going final with our test standards and test guidelines, that office and ours can
merge efforts to come up with protocols that are identical to the maximum extent
allowable under the two different laws. In addition, we are also working with the
Interagency Regulatory Liaison Group, to make sure that there is maximum
consistency not only across EPA offices in defining acceptable testing methodologies,
but also across the federal regulatory agencies.
The other section of TSCA in which testing comes into play is Section 5, which
applies to new chemical substances, new chemical substances being those that are
not contained on the TSCA chemical inventory of all existing chemicals in commercial
use. The inventory has been published. It is being distributed now, and Federal
Register notices are going out indicating the official date of the inventory publication to
be June 1. On July 1, thirty days after publication, the premanufacturing notification
review process will be initiated.
The premanufacture notification review (mandated under Section 5) will consist
of evaluating certain information about the chemical's identity, use, production,
exposure, etc., and also all available health and safety studies on the material that is
being proposed for manufacture. We do not generally have the authority under
Section 5 to require testing. However, testing of new chemicals can be required if they
fall under Section 4, in particular, if they fall in categories of chemicals that might be
specified under Section 4. Also, under Section 5(e), the EPA has the authority to issue
an order to limit production, distribution, or use as necessary if there is insufficient
information provided to the EPA to enable it to determine whether the chemical poses
an unreasonable risk, or if the EPA can demonstrate that the chemical may pose an
unreasonable risk, or if the chemical is produced in substantial quantities with
substantial release or exposure.
Because of the Section 5(e) provision in which the EPA may limit exposure or
production in some fashion and because of concerns by industry and others as to
what the EPA expects and will be satisfied with, we have contemplated coming forth
with Section 5 testing guidance, which would describe the type of testing and the
testing scheme or approach we would most like to see performed. To date, this has
been a very controversial area, as I am sure many of you are well aware. We have had a
number of open public meetings and have decided to move along fairly deliberately in
this area before we propose the document. As a result we decided that rather than
publish a formal proposal, we would first publish a discussion document which laid out
the various alternative approaches available to the EPA for testing and then suggest a
menu of tests we thought might be appropriate to incorporate into a future testing
guidance. Also, we want to solicit as much public comment as possible regarding both
-------
the scientific merits of the tests and the various arguments, scientific and other, which
might impinge upon the choice the EPA would make among the various available
alternative testing approaches. That discussion document was published March 16 in
the Federal Register. It is still open for comment, and we are getting a large response.
With respect to assessments, our office is involved in four different types. One is
under the chemical substances premanufacture notification process. Another is a
review of chemicals to see if they meet the Section 4 criteria for testing, and two others
deal with existing chemicals. These last two include our comprehensive assessment
process and a priority problem assessment, the priority problem assessment being a
subset of comprehensive assessment and a faster track. These latter two are
multistaged assessments of existing chemical substances, multistaged primarily so
that we can take the limited amount of resources available in our program and target
them as efficiently as possible on the highest priority problem chemicals. This focuses
our limited energies on those chemicals of greatest importance. Therefore, we do
basically a quick and dirty analysis on a very large number of chemicals and then
narrow down to what in the last analysis are many fewer chemicals.
In the later stages of chemical assessment, we go through all the relevant
scientific information in an effort to pull together as good a risk evaluation as possible
and use this evaluation to base actions or potential actions under TSCA or other
regulatory statutes. Because of both our testing authorities and our assessment
activities, we are particularly interested in evaluating the state of the art of various
testing approaches, in particular today, to focus upon subchronic toxicity testing. We
are very interested in the state of the art of this type of testing method in the context of
TSCA authorities and of our ability to use it in terms of both our testing schemes and
our evaluations of information.
We are particularly interested in this workshop and addressing questions such as
proper protocol designs and the extent to which protocol design should be very
specific or more general. We recognize that under Section 4 of TSCA, the EPA has to
specify beforehand what testing should be done and is not able to turn away
inadequate studies later, as long as the studies meet the required specifications. In this
regard, the EPA is unlike the Food and Drug Administration and the Office of
Pesticides Programs, which register materials only after acceptable studies are
performed.
Further, we are interested in any judgments the workshop can make with respect
to the relevance of the subchronic toxicity test results for risk assessments. We are
also interested in the judgments of this particular group with respect to the
relationship of subchronic toxicity tests to other types of tests, such as acute tests or
chronic tests in an evaluation scheme. For example, Dr. Seifter, my Medical Science
Advisor, is of the opinion that perhaps a subchronic toxicity test does not have a place
in a TSCA evaluation scheme at all, tha.t we may be able to substitute a shorter test of
14 or 28 d with considerably more information on metabolism, biochemistry, etc., and
then, depending upon the results, follow directly with a chronic test. We would be
interested in the views of this group on this idea and others.
I presume that there will be ample opportunity to ventilate that particular
proposal. We are, therefore, interested in these types of activities, and you can
understand why this particular workshop is so important to our office. Because of our
-------
interest we asked the Oak Ridge National Laboratory to assist us in establishing this
workshop, and we are very pleased with the group we have been able to assemble
here to meet this week. We deeply appreciate the sacrifices each and everyone of you
has made in your busy schedule to come out here, roll up your sleeves, and work with
us on this agenda. Thank you.
-------
REVIEW OF
CURRENT ACTIVITIES IN
TEST STANDARDS DEVELOPMENT
NORBERT P. PAGE
U.S. Environmental Protection Agency
During the last few years, increased legislative responsibilities have been
assigned to government agencies in both the U.S. and abroad to better evaluate
environmental chemicals for their potential toxic hazards. Safety assessment has
been routine in the pharmaceutical industry for years. Safety testing of food
additives and pesticides and, to a lesser extent, industrial chemicals has also been
conducted. Now virtually all chemical manufacturers are responsible for developing
data to demonstrate the toxic potential of their products, regardless of their
products' intended uses.
Many laws have been passed which require that the government agencies
develop standards or guidelines by which testing must be conducted. With these
new demands, a flurry of activity has taken place both in the U.S. and abroad in
the test standards or guidelines area. Dr. Muir has already spoken to the specific
needs of the EPA in fulfilling its mandates under the Toxic Substances Control
Act.
It is my intent in this presentation to demonstrate how the EPA relates to
other ongoing activities in the testing area. Four different acts within the purview of
the EPA alone require the testing of chemicals and the development of testing
guidelines. These are the Toxic Substances Control Act (TSCA), the Federal
Insecticide, Fungicide, and Rodenticide Act (FIFRA), the Clean Air Act as
amended in 1977 (fuels and fuel additives), and the Resources Conservation and
Recovery Act (hazardous wastes).
Among the other regulatory agencies, the Consumer Product Safety
Commission (CPSC), through the National Academy of Sciences (NAS), revised
its procedures for toxicity testing of household products (NAS 1977). This was
completed about two years ago. The Food and Drug Administration (FDA) is now
engaged in a cyclic review of approximately 2100 direct food additives. They are
developing procedures for toxicity assessment including identification of test
procedures. The Food Safety Council (1978) has also produced a series of test
procedures. In an FDA-related activity, the Pharmaceutical Manufacturing
Association has drafted its Preclinical Guidelines for Assessment of Drug and
Medical Device Safety in animals.
-------
On the international scene, several foreign countries, the World Health
Organization (WHO), the European Economic Commission (EEC), and the
Organization for Economic Cooperation and Development (OECD) have been
producing test guidelines. I am sure I left out a few other ongoing efforts. Many of
you have participated in these guideline activities.
One could pick from any number of protocols, and having such a wide range
of choices may at first seem to be an advantage. On the other hand, conflicting
methods may be required of the manufacturer of a product—especially where
there is more than one use—to satisfy the requirements of several agencies or
countries.
Attempts at Harmonization
What is being done to harmonize guidelines? Harmonize—that is the new
word about town that sounds better than, but means about the same as,
standardization.
Within the EPA, the Office of Pesticides Programs and the Office of Toxic
Substances have joined together to develop a common set of guidelines. Those
implementing the Clean Air Act in the testing of fuels and fuel additives are
committed to using the TSCA/FIFRA guidelines rather than producing their own.
Thus, the EPA will basically have a single set of toxicity testing guidelines.
Within the federal bureaucracy, the regulatory agencies have banded together
to come up with consistent and cooperative regulatory actions. To achieve this,
the Interagency Regulatory Liaison Group (IRLG) was formed. One work group
was chartered specifically for toxicity guidelines and standards. Five federal
regulatory agencies participate: EPA, FDA, CPSC, the Occupational Safety and
Health Administration, and the U.S. Department of Agriculture. Dr. Victor
Morgenroth (FDA), the chairman, is participating in this conference. When IRLG
develops final guidelines, the agencies will adopt them.
As far as the International circuit is concerned, the U.S. is an active
participant in both the WHO and OECD programs. Of the two, the OECD will
likely carry the most clout. Participating countries will likely accept the methods
developed and incorporate them into their chemical control laws.
Status of Activities/Concept of Progress
Last August under FIFRA, the EPA proposed pesticide registration guidelines
for the major health effects. On May 9, the EPA proposed chronic standards and
Good Laboratory Practices (GLPs) under TSCA. It is expected that in July or
August most of the remaining TSCA standards for health effects testing will be
proposed. These will be essentially the same as the FIFRA guidelines. By the end
of the year or in early 1980, joint TSCA/FIFRA guidelines should be ready for
finalization in the Code of Federal Regulations. The IRLG is also participating with
the EPA in reviewing public comments and revising the proposed guidelines.
The IRLG will soon propose a series of guidelines for acute effects and
teratogenicity. It is expected that subchronic guidelines will be ready for agency
reviews by late summer and those for chronic tests, by fall.
-------
The OECD has formed several groups of experts to prepare universally
acceptable guidelines. The U.S. has the lead in establishing guidelines for chronic
effects and GLPs, the British, for short-term effects, and Sweden, for
step-systems. Other groups are also working in ecotoxicity, biodegradation,
bioaccumulation, etc. Proposed guidelines are now prepared for eye and skin
irritation, acute and subacute toxicity, and teratology. As with the IRLG, it will be
summer before the OECD has drafted a subchronic test protocol.
Review of Recently Proposed Subchronic Toxicity Tests
Table 1 lists the basic experimental design for four of the recent activities that I
mentioned. There are no great surprises here, as the designs are similar to those that
have been in use for several years. They all call for a fairly detailed clinical and
pathological examination, including microscopic examinations. There are some
differences in organs to be weighed, tissues to be examined, etc. These differences
should be reviewed by the appropriate work groups. Whereas there is fair consistency
in certain design aspects, other areas are controversial (e.g., duration, species). The
FDA's draft for food additives testing is very similar to that for the IRLG shown in
Table 1.
In addition to the 90-d oral subchronic tests, other subacute or subchronic tests
have recently been proposed. The FIFRA guidelines provide a separate protocol for a
90-d subchronic inhalation toxicity test. The basic design is similar to that presented
for the 90-d oral test, except that only the rat is required, and the number per group is
only 10 instead of 20. The remainder of the design is virtually the same but with
increased clinical and pathology examination of the respiratory system. The IRLG has
not yet prepared guidelines for a subchronic test by the inhalation route.
The proposed FIFRA guidelines and the IRLG guidelines also include dermal
toxicity tests of 90-d durations. In both, albino rabbits are the species of choice. Eight
to ten young adults of both sexes are used for each of three dose levels plus controls.
The highest dose is to be selected to produce toxic effects but not cause excessive
mortality (over 10%). The lowest dose should be a no-observed-effect level.
The FIFRA proposed guidelines provide a 21-d dermal toxicity test for products
whose pesticidal use is likely to result in repeated human skin contact. This is not
needed when a 90-d dermal toxicity test is performed. Except for duration, the
protocol is the same as for the 90-d dermal test.
The FIFRA proposed guidelines require a delayed neurotoxicity test with
mammals if neurotoxicity was observed with acute tests. The species to be tested is
that which demonstrated the neurotoxicity signs. Both sexes are to be used for
mammals, whereas only the female domestic chicken is used. At least three dose
levels each with 10 animals of each sex are employed with exposure for a 90-d period.
The highest dose is one that induces toxic effects, whereas the lowest should not
induce any deleterious effect. Observations for behavioral abnormalities, including
locomotor activity, are called for, along with very detailed and complete pathological
examination of the central nervous system and peripheral nerves.
-------
o
(f>
ts
s
^
'u
'*
o
-M
s
hi
O
u
o
X
u
.c
3
rccenl
,0
c
0)
'i
•o
3
'c
1
X
u
I-H
_w
2
£
r*
u
c
a
o
U_
>°°
Jj o
*n3 C^
to
T3
O
£
So
r-~
.c 0;
0) c
1 J
*o 15
"C N
-S 'c
3J a
5- g,
O
^u
O ^
12
01 — '
£ §
> g
£5
OT C
S S
S a
£ 3
S
C -o^
"S o>
11
Q,. o^
LL, ^_^
E
fO 0) _L
§ ^ " * 2 ^ - 1 S
.S 1 -si £ i^»n
:= -*- >> , to-^i-fch:
g T, T ^7 s ^ -8 ~ - E >
•£ S 2 « S w -^ai o~o3~"Sf3
1- s 7c Tc vz §^a)T3
^ = -e Jj -S ^ J. g J. e g 's g
i. « c co co g f, ^^Sajg
o^ -c Z •% c -§c x = | -a g "a §
3>-oO go oo S1 -° greS-Cj:
H mz az az mi 3
QJ
c S
2 co £
C )_ U C
"O O tl Q> O
r= i. w U -i- ~
§ w c s w t;
i: fa +- ^ ro
"r- ro ti £ "TT; o QJ
g -a, 8 „ "a ^ i >-
•g^g SS S.S |^|2-
!! * 11 Ji !§ Is
^^ora^o ^d 21-' o^
HrTi (V zi _ J _». "T1 i
P-J U- <— i I-H -^f -L, , — 1
J^/)
_C
C .,
O O
p £ _ >
| 1o°V g£g I S =, i
(u ^ -+-. — ' " — C -•— ' ^v C
2 ^7 7S7 til ^141
^ ^ 1 a i a 111 -II 1 1 §
il ,S3o3 ^rT 21 -^ E g -
H CQCCQCCQO coJ. _J
r ">
^ > fi c
§ 0 „ J* ^
J= P -C +_ a; >\ =r
*- u) *" "U •*-• u i- "*"* ^
•s 1 o s i =s § i o 3.
ra~ 3° |6
-------
The 28-d Toxicity Test
The FDA is proposing a 28-d toxicity test that they refer to as a range-finding oral
toxicity study. It is intended to determine the target organs and find a range of doses
for subchronic or chronic tests. It is similar to the 90-d IRLG oral subchronic test,
except for a 28-d exposure and fewer animals per group (i.e., 10 rats instead of 20 and
2 dogs instead of 4 or 6).
The OECD/EEC has also proposed a 28-d test using the rat only, but it is
otherwise like the FDA test. The 28-d test is to be used in a step-sequence system. The
concept of the step-sequence scheme is to determine the tests to be performed based
upon the potential for exposure and effects. The current sequence for testing new
chemicals as proposed by the EEC and OECD is as follows: for less than one ton, no
tests are needed. For all chemicals produced in quantities of one ton or more, a series
consisting of acute toxicity tests, a battery of two or three mutagenicity tests, and the
28-d test are required. For those chemicals to be produced in quantities greater than
10 tons, more testing is needed (e.g., fertility test, teratology tests in one species, more
mutagenicity tests, and a 90-d subchronic test). At production of 1000 tons, the ante
goes up even further in that chronic tests and pharmacokinetic studies would be
undertaken.
I hope from this quick review you can see the role that the subchronic test will
play in the U.S. and international plans for safety assessment of chemicals. The
subchronic test, more than any other, plays a pivotal or key role in virtually all safety
assessment schemes. Much of the driving force behind protocol design by OECD and
TSCA for new chemicals is that of economics. We recognize that testing costs can
stifle the introduction of new chemicals and that inconsistent testing requirements can
create international trade barriers. These are both issues of concern to EPA in its
implementation of TSCA. This workshop needs to examine the trade-offs in
sensitivity and usefulness of the subchronic test if we reduce requirements (e.g., 2
species to one, 90 d to 28 or less, reducing clinical or pathology requirements, etc.).
We need solid data to predict the consequences of such modifications in test design.
REFERENCES
Federal Register. 1978. Proposed Guidelines for Registering Pesticides in the U S. Hazard
Evaluation Humans and Domestic Animals U.S. Environmental Protection Agency (40 CFR 163).
43(163) 37336-37403
Food Safety Council 1978. Subchronic Toxicity Studies. In. Proposed System for Food Safety
Assessment Columbia, Md pp. 83-92.
Interagency Regulatory Liaison Group. 1979a IRLG Guidelines for Subchronic Ingestion Testing
(unpublished)
Interagency Regulatory Liaison Group. 1979b. IRLG Guidelines for Subchronic Dermal Toxicity
Tests (unpublished).
National Academy of Sciences Committee for the Revision of NAS Publ. 1138 1977 Principles
and Procedures for Evaluating the Toxicity of Household Substances NAS, Washington, D.C. 130 pp.
Page, N P 1977 Concepts of a Bioassay Program. In: Environmental Carcmogenesis. H. Kraybill
and M. Mehlman, eds Hemisphere Publishers Inc , Washington, D C pp. 87-171
Sontag, J. M., N P Page, and U Saffiotti 1976. Guidelines for Carcinogen Bioassay in Small
Rodents Department of Health, Education, and Welfare Publ. No. (N1H) 76-801. 65 pp.
World Health Organization 1978. Acute, Subacute, and Chronic Toxicity Tests In- Principles and
Methods for Evaluating the Toxicity of Chemicals Part 1. Environmental Health Criteria 6. Geneva, pp.
95-115
-------
-------
CHARGE TO THE WORKSHOP
AND THE INDIVIDUAL COMMITTEES
WAYNE M. GALBRAITH
U.S. Environmental Protection Agency
A subchronic toxicity study can be defined as qualitative and quantitative
evaluation of the physiological and pathological changes in the rodent and/or
nonrodent resulting from regularly repeated exposure to a chemical over a period of
one to three months.
It is generally recognized that toxicity resulting from acute chemical exposure
and that resulting from subchronic and/or chronic exposure may differ. The results
of acute studies may not be adequate to predict toxicity which results from
repeated low-dose chemical exposure. Therefore, the subchronic toxicity study is
fundamental in determining the toxicity profile of chemical substances to which
repeated or continuous human exposure can occur.
It is the objective of this workshop to summarize the various factors which
should be incorporated into the experimental design of an "end result"1 subchronic
toxicity evaluation and to examine the scientific basis for these factors. The
ultimate objective of the workshop is to provide the rationale for cost-effective and
scientifically sound subchronic toxicity testing procedures.
The committees should summarize the historical data bases which justify the
inclusion of the various parameters in a subchronic study. When it is determined
that expert opinion cannot be substantiated by an adequate data base, research
recommendations should be made. In the course of your deliberations, all
committees are also requested to devote attention to defining the technical
terminology used in the committee documents.
It is not the objective of this workshop to duplicate what has been accomplished
in a number of reviews and proposed test systems published in recent years (National
Academy of Sciences 1975, 1977; World Health Organization 1978; Food Safety
Council 1978; U.S. Department of Health, Education, and Welfare 1971; Association
of Food and Drug Officials of the United States 1959). Nevertheless, some
duplications of previous efforts are unavoidable and desirable to the end that the
workshop product is comprehensive. When appropriate, the committees are
encouraged to summarize and update pertinent information in previous reviews.
'The term "end result" is intended to distinguish a study designed as the terminal toxicologic
evaluation of the toxic effects of a chemical from a subchronic study used to determine dose levels for a
chronic study.
-------
12
The workshop should focus on the testing of chemicals covered by the
Federal Insecticide, Fungicide, and Rodenticide Act and the Toxic Substances
Control Act. The workshop is divided into committees which will address five
general topics pertinent to subchronic toxicity evaluation.
The first committee will discuss general experimental design. Considerations
such as minimum test duration,- dosage, number of animals, test species, and
animal age at time of chemical exposure will be evaluated by this group. For
example, many scientists have difficulty agreeing with the requirement of a
maximum permitted mortality in the high dose group in a subchronic study. Is this
requirement justified? Would it be preferable to employ 10-15 rats per sex/dose
group and have 5 or 6 dose levels rather than 3 dose levels with 20 rats per
sex/dose group? What are the advantages and disadvantages of each study
design?
The second committee will review parameters evaluated in a subchronic
toxicity study. This includes general observations, clinical laboratory tests, and
pathologic evaluations. A suggested minimum base set of general observations for
rodent and nonrodent subchronic toxicity studies should be developed. It is also
requested that the second committee address the desired frequency of chemistry
and hematologic evaluation.
It has been suggested to the Agency that total serum protein, serum albumin,
and serum protein electrophoresis are a better monitor of liver toxicity than serum
glutamic oxaloacetic transaminase and serum glutamic pyruvic transaminase. Is
there an adequate data base substantiating this statement? Are cholesterol and
uric acid determinations justified in rodent and nonrodent subchronic toxicity
studies? Should sperm motility be evaluated in control and treatment groups at the
time of sacrifice? Are there other tests not routinely performed which should be?
Are there special techniques, histochemical or other, which can be employed to
increase the sensitivity of a subchronic toxicity study for detecting neurotoxicity?
What comprises an adequate neurologic examination in the rat and dog in a
subchronic toxicity study?
It is recognized that no toxicity testing protocol will be adequate in every
experiment. For this reason the third committee has been asked to evaluate the
relationship of protocol design to chemical class and characteristics. What
observations, clinical laboratory tests, and/or pathologic techniques not routinely
made should be included in protocol on the basis of chemical class? That is, what
modifications in protocol design should be made on the basis of the
physical-chemical properties of the test substance? What are the hazards or pitfalls
associated with the subchronic evaluation of specific chemical classes? Examples
that readily come to mind are the need to evaluate methemoglobin when testing
nitrates and nitrites, the need for a cholinesterase activity evaluation when testing
an organic phosphate or carbamate pesticide, and the potential aspiration hazard
when conducting an oral-dosing study with certain hydrocarbons.
The fourth committee will evaluate the criteria to be employed for
extrapolations from one route (i.e., oral, dermal, inhalation) to another route of
administration. Because the cost of a subchronic inhalation study is at least twice
that of a comparable oral-dosing study, it is imperative to know when inhalation
-------
13
studies should be performed and when studies employing other dosing routes will
be adequate. In this regard, what role should the results of acute inhalation and
oral-dosing studies play in determining the need for subchronic studies using more
than one route of administration? What criteria will be used to extrapolate from
oral-dosing studies to inhalation studies? What assumptions must be made? What
other studies, if any, must complement oral subchronic studies in order to make an
extrapolation?
The fifth committee will evaluate the limitations of subchronic toxicity studies.
What subchronic findings should trigger the requirement for a chronic study?
What manifestations of toxicity are not likely to be observed in a subchronic study
but may be found in a chronic study? How much information regarding maximum
tolerance levels for humans can be obtained from a subchronic study? Is
comparative metabolism the key factor in evaluating a toxicity profile of the same
chemical in the rodent and nonrodent? If this is the case, is it possible to eliminate
the need for a complete study in one species when it is determined that the
adsorption, distribution, metabolism, and elimination pattern is similar for the
rodent and nonrodent? How useful are the results of acute toxicology studies for
determining if a metabolic difference exists in the rodent and nonrodent?
Other potential topics for consideration by this committee are the desirability
of performing combination tests (e.g., subchronic and teratology in the same
study), dosage extrapolation from laboratory animals to man, and results of acute
toxicity which should trigger subchronic studies.
By Thursday morning each committee should have prepared draft documents
which address the assigned topics. The document may also contain research
recommendations and discussions of additional problems which the participants
consider pertinent to subchronic toxicity testing. Of course, because of time
constraints, it is recognized that all of the suggested topics may not be discussed.
Following the workshop, each participant is requested to make additions,
corrections, or other changes in his or her committee's draft document and
forward them to the appropriate chairman.
REFERENCES
Association of Food and Drug Officials of the United States, 1959 Appraisal of the Safety of
Chemicals in Foods, Drugs, and Cosmetics. 106 pp.
Food Safety Council. 1978. Proposed System for Food Safety Assessment Columbia, Md. 136 pp
National Academy of Sciences 1975 Principles for Evaluating Chemicals in the Environment.
National Academy of Sciences. Washington, D.C 454 pp.
National Academy of Sciences. 1977 Committee for the Revision of NAS Publ. 1138 Principles
and Procedures for Evaluating the Toxicity of Household Substances NAS, Washington, D C 130 pp.
U.S. Department of Health, Education, and Welfare. 1971 Food and Drug Administration
Introduction to Total Drug Quality U.S. Government Printing Office, Washington, D.C No 1712-0020.
World Health Organization. 1978. Acute, Subacute, and Chronic Toxicity Tests In Principles and
Methods for Evaluating the Toxicity of Chemicals Part 1 Environmental Health Criteria 6. Geneva pp
95-115.
-------
-------
PROCEEDINGS OF THE WORKSHOP ON
SUBCHRONIC TOXICITY TESTING
1. INTRODUCTION
1.1 Scope
Under the Toxic Substances Control Act (TSCA) and the Federal Insecticide,
Fungicide, and Rodenticide Act (FIFRA), the Environmental Protection Agency
(EPA) is required to provide standards or guidelines for the testing of chemicals for
health and environmental effects. The EPA is also required to perform hazard or risk
assessments for selected existing chemicals in the environment as well as for all new
chemicals to be manufactured in the United States or to be imported. In addition to
EPA's legislative requirements, other U.S. federal regulatory agencies and other
countries have similar requirements to provide testing guidelines or guidance. The
EPA has joined with other U.S. agencies in forming the Interagency Regulatory
Liaison Group (IRLG) and with other countries in the Organization of Economic
Cooperation and Development (OECD). Both the IRLG and the OECD have work
groups to design testing procedures for the development of data for risk assessment
of chemicals.
The subchronic study is a key test in assessing the relative safety of chemicals
to which repeated or continuous exposure to humans can occur. This test is likely to
be a fundamental part of any guidelines for the toxicity testing of new chemicals
entering the market.
1.2 General Principles
Subchronic toxicity procedures are designed to determine the adverse effects
that may occur with repeated exposure over a part of the average life span of an
experimental animal. The boundary between the subchronic and chronic dosing
regimes is often taken as 10% of an animal's lifetime; thus subchronic will include
dosing regimes described as subacute (Organization for Economic Cooperation and
Development 1979). Subchronic studies often involve multiple dosage levels, some of
which may exaggerate by significant multiples the anticipated levels of human
exposure. Testing procedures normally include the routes of exposure expected for
man. Usually one, two, or more animal species are exposed to the chemical for a
period of 90 d to evaluate the potential hazard. In the United States, one rodent (the
rat) and one nonrodent species (the dog) are generally employed as the two species. In
some other countries, the mouse, rather than a nonrodent, is used as a second
species. The selection of animal species for subchronic studies should consider,
15
-------
16
whenever possible, whether the test species' metabolic handling of the chemical or its
analogs is similar to that of humans. Exposure levels used in the studies are lower than
those in the acute toxicity procedures, and lethality is usually not an end point.
Instead, by including many types of observations (behavioral, growth rate,
hematological, biochemical, and functional) and extensive pathological evaluations,
more subtle types of adverse effects can be detected.
Man may be exposed for the majority of his lifetime to low levels of a wide variety
of chemicals. Usually the degree of exposure is insufficient to produce overt signs of
toxicity; thus cause-effect relationships cannot be easily established. Epidemiological
studies may assist in this respect, but because man is often exposed simultaneously to
several chemicals, it is difficult to establish unequivocally the degree of hazard
associated with any one chemical.
1.3 Objectives
For purposes of this report, the subchronic toxicity study is defined as an "end
result" study, that is, it will be the only long-term study available to the regulatory
agency responsible for making decisions about the risks of human exposure to a given
chemical. Therefore, the subchronic study should define target organs and tissues in a
manner sufficient to provide the basis for risk estimation. The studies should provide
(a) the demonstration, with appropriate methods, that the toxic effects observed at
the high-dose level are not observed at least at the low-dose level and (b) the
delineation of tests that would also evaluate potential effects from long-term exposure
conditions. The results from the subchronic study should be applicable to potential
exposures of longer duration; however, it is not expected to be all-encompassing and
predictive of such effects as carcinogenicity, teratogenicity, certain reproductive
effects, or effects on longevity.
The workshop objectives included
• critically examining the subchronic study as it might be applied to general chemical
testing programs and evaluating the effect of variations in study parameters on
assessment potential;
• identifying the deficiencies in the existing scientific knowledge and recommending
research to strengthen the testing program; and
• identifying the related scientific issues to be resolved in future workshops or
experimental programs.
2. RELATIONSHIP OF PROTOCOL DESIGN TO
CHEMICAL CHARACTERISTICS
The experimental design is based on the utilization of information available from
physical and chemical data, acute studies, and structure-activity analogies with
known substances. Variations in subchronic toxicological evaluations may be
indicated for different chemical classes and characteristics. However, current
knowledge does not permit development of a specific and definitive test design for
each identifiable chemical class.
-------
17
The outline design is based on the fundamental premise that a subchronic study
is rarely conducted in the absence of prior information on the test chemical. The
physical and chemical data listed in Table 2 will usually be known for the test
substance. In addition, the investigator will usually have an indication of the acute
toxicological response. It is this information (Table 3) considered together with the
physical and chemical properties and, when available, the biological activity of
analogous structures that will determine the design of the subchronic study. Other
experimentally derived data regarding biological disposition of the toxicant, if
available, should be utilized.
Table 2. Chemical and physical information
Molecular weight
Molecular formula
Structural formula
Electrophile/neutrophile
Purity
Acidity/alkalinity
Particle characteristics °
Density
Corrosivity
Solubility in H2O
Lipid solubility
Melting point0
Boiling point"
Vapor pressure"
Dissociation constant (pK)
Stability at various pH's
Stability to heat and light
Relative to route of administration.
Table 3. Acute toxicity information
Time to onset of signs as a function of dose
Time to recovery
Weight changes
Organ effects
Observation of skin, fur, eyes, mucous membranes, etc.
Changes in behavior
Signs of autonomic nervous system effect, such as tearing,
salivation
Changes in respiratory rate; depth and signs of respiratory
irritation
Cardiovascular changes, such as flushing
Central nervous system changes, such as tremors, convulsion,
coma
Observation of dropping pans for evidence of eating or
noneating, urine color and volume, diarrhea
Time of death
Necropsy findings
Slope of the dose response line
-------
18
The information depicted in Fig. 1 allows the investigators to make one of the
following choices: (1) adequate data are available to predict hazard and a subchronic
test is not necessary; (2) adequate data are not available to predict hazard and a
routine subchronic test is needed; (3) the data indicate that because certain effects are
expected, a modified subchronic study utilizing an expanded protocol with
nonroutine tests should be designed; and (4) the data indicate the need for a chronic
test that must be designed through use of a modified subchronic test. In some cases
additional information may be available to indicate that no more work is necessary.
This information may include available experimental data and such factors as the use,
distribution, and production volume of the compound, as well as the likelihood of
human or environmental exposure.
Structure-activity correlations can sometimes provide a priori information
concerning biological activity related to the molecular structure of the substance
under consideration. This information can be used to make either quantitative or
ORNL-DWG 80-18344
Intended production
volume, exposure
potential, use, etc.
Acule toxicity
(see Table 3)
Data
input
Decision
level 1
Decision
level 2
Fig. 1. Selection process for designing toxicity studies.
-------
19
qualitative estimates of biological activity. The examination of structurally related
chemicals of known toxicity which have physical-chemical properties (such as
partition coefficients or Taft-Hammett constants for steric and electronic properties
of substituents) similar to those of the test chemical may be useful in estimating the
biological activity of that test chemical. This procedure may be of use either with a
homologous series of chemicals or with chemicals that act on similar molecular
targets. However, the necessary information for estimating biological activity may not
always be available for many chemicals.
Some of the qualitative structure-activity correlations for the industrial chemicals
of interest may relate to the carcinogenic potential rather than the general toxic or
metabolic activity of the molecule. However, in the absence of experimental data,
qualitative structure-activity correlations have some use in the prediction of potential
target-organ toxicity or metabolic activity. Structure-activity correlations, therefore,
may be of assistance in selecting toxicological parameters for evaluation.
In conclusion, much reliance in designing a subchronic study can and should be
placed on the physical-chemical profile of the test compound and its intended use.
Biological data from acute studies can also provide some guidance. Data on uptake,
distribution, and elimination of a compound are dependent upon the availability of
appropriate analytical methodology. Structure-activity relationships should be viewed
and applied with caution and should never totally substitute for actual experimental
data.
3. EXPERIMENTAL DESIGN
3.1 Exposure and Duration
Routes of entry for the chemicals of concern are inhalation, ingestion, and skin
exposure. Ideally, the subchronic study should be conducted by the route of major
concern relative to human exposure. It is recognized, however, that the subchronic
study itself would generally be conducted by one route of administration and that
multiple factors may be involved in the selection of that route.
An important concept in subchronic toxicity testing is that repeated doses of a
substance, when adequate doses are administered, are likely to exert most of their
potential long-term effects within 90 d. A lifetime "no-effect" dose can usually be
predicted from such studies, excepting those effects referred to in the introduction.
Weil and McCollister (1963) assessed the predictability of lifetime no-effect doses from
subchronic data. Of 22 chemicals none had a long-term no-effect dose that was less
than 1/20 of its 90-d no-effect dose. For most chemicals the ratio of the 90-d no-effect
dose and the long-term no-effect dose was less than 10:1. The review of McNamara
(1976) also supports these conclusions. Durations of less than 90-d have also been
evaluated. Weil et al. (1969) found that even 7-d test results might be a sufficient
indicator of long-term toxicity if used with a safety factor of appropriate magnitude.
However, few references in the literature assess clinical and pathological effects for
the shorter term with the 90-d studies.
In the majority of cases, per oral and inhalation studies should be accomplished in
3 months. In most inhalation studies comparable results will be obtained from either
-------
20
the 5- or 7-d/week exposures. The inhalation exposure schedule should be5 d/week,
6 h/d. For repeated dermal route studies, a maximum of 21 applications to rabbits on a
5-d/week basis is considered to be practical and of sufficient duration because of
irritation and stress from repeated applications, daily restraint, and possible
intercurrent infections.
3.2 Dose
To establish the extent of the toxic response, the highest dose should provide a
distinct toxic effect but not lethality, whereas the lowest dose should produce no
detectable toxic effect. To obtain maximum information on the dose response
characteristics of the chemical, at least one intermediate dose is generally used.
3.2.1 Selection of doses
Guidance on the selection of doses for subchronic toxicity studies may be
obtained from the results of prior acute and repeated high-dose studies. Assuming
that the LD50 of the chemical has been established, some fraction of that dose may be
assumed to be nonlethal but still toxic over the test period. This dose usually falls
between 10 and 25% of the LD50, depending largely on the slope of the acute dose
response curve.
For chemicals having a tendency for bioaccumulation, the selection of dose levels
is particularly difficult. Pharmacokinetic studies may assist in selecting the dose levels,
because the elimination pattern may provide guidance on the extent of possible
bioaccumulation. However, many of these factors may be unknown when the study is
designed, so alternative approaches have been utilized for this purpose. Preliminary
range-finding tests using 5 or more groups of rodents (perhaps as few as 2 or 3 animals
per group) for a period of 3-4 weeks, with doses at twofold to fourfold intervals can be
used to select dose levels.
The high-dose level should produce some significant toxicological effect without
a high incidence of mortality that would prevent meaningful evaluation. The
successful selection of a high dose that will produce these adverse but nonlethal
effects is partially related to the results obtained from these studies.
3.2.2 Limitations of maximum levels
Occasionally, toxicological effects are not observed with a chemical even when
very high doses are administered in a subchronic study. In such cases, there are
maximum dose levels that need rarely be exceeded. In these studies, the maximum
dose levels should not cause nutritional imbalance and/or affect the defense
mechanisms. Maximum dose levels are considered to be 5% of the diet or 2.5 g/kg of
body weight per day for peroral administration, 2 g/kg of body weight per day for
dermal exposures, and 20 mg/m3 (6 h/d) for inhalation of particulates, gases, and
vapors.
Whenever these maximum dose levels are to be exceeded in actual use by
humans, levels higher than these maxima should be used in subchronic studies, and
care should be taken to avoid complications from artifacts that would prevent
-------
21
meaningful utilization of the data. In some cases, factors such as low-vapor pressure,
explosive limits, inability to achieve chamber concentrations, and insolubility and
instability in vehicles may further limit the practical attainment of the desired high-
dose levels. In a few cases, a chemical with low acute toxicity intended for low-volume
productions may not be available in sufficient quantity to provide the desired high-
dose levels in a subchronic study.
Because in this case the desirable objective of the subchronic toxicity test is to
establish dose response patterns and no-observed adverse-effect levels, a minimum of
three dose levels including at least one intermediate dose is generally used.
3.3 Species Selection
The animal species whose biological processing of the material being tested is
most similar to man's should be used for subchronic experimentation. Because such
information may be unavailable when subchronic studies are initiated, it is suggested
that at least the rat, and probably the dog, be employed. Because in some instances
the rat's sensitivity has been equal to or greater than the dog's, the rat has been at least
as useful as the dog in detecting the lowest level at which an effect has been observed
(Avidao 1978). However, the dog has shown greater sensitivity in other instances
(Litchfield 1961; Homan 1972). In special circumstances, with a small number of
chemicals, other species have been shown to be more sensitive than either the rat or
the dog (Smith 1979). However, the use of additional species would often not
appreciably affect the determination of a safe exposure level for man.
3.4 Number of Animals
The number of animals to employ in subchronic tests varies depending on the
goals and needs of the investigator. The participants disagreed concerning the
appropriate level of statistical analysis and the value of clinical chemistry/hematology
tests. Therefore, two sets of animal numbers were suggested. If biochemistry and
hematology tests are to be used as a major part of the assessment of toxic effects, then
20 rats per sex per level and 5 dogs per sex per level are necessary. If less use is made
of clinical chemistry tests, then the numbers suggested are 10 rats per sex per level
and 3 dogs per sex per level. For repeated skin application studies, 5 rabbits per sex
per dosage level are recommended. Increasing the number of animals per level above
that recommended will markedly increase the cost and decrease the efficient
utilization of facilities and personnel; therefore, the choice of 3 dogs and/or 10 rats, or
5 dogs and/or 20 rats, should be justified by the investigator.
3.5 Age of Animals
In selecting the age of animals for use in a subchronic toxicity study, the ideal
situation would have all age groups represented. Use of the neonate or fetus would
allow the evaluation of toxic effects of chemicals on the earliest life stages. It would
also be valuable to assess the toxic effects in older animals near the maximum life
span. This would provide an estimation of the potential toxic effects for all ages of the
human population. However, this ideal situation cannot easily be achieved because of
such practical considerations as the needs to establish baseline data, allow for
-------
22
acclimatization to the housing conditions, ensure the absence of nontreatment
diseases, and even acquire test animals. All these require time and prevent the use of
extremely young animals. Interference from normal aging effects and the difficulty in
obtaining sufficient numbers of older animals of the same age make the use of older
animals very difficult also. The subchronic test cannot be used to evaluate specific
age-related effects either in the fetus, neonate, or older animals. Special studies would
be required for such data. Therefore, in routine testing it is generally recommended
that rodents be started on treatment as young as practically possible. This
conceivably could be from 3 to 4 weeks of age; however, a period of 2 to 4 weeks
usually is needed to assess the health and maturity of the animals and to perform
pretest procedures. Thus, the earliest practical age to start rodents on test is 5 to 8
weeks.
For nonrodents, especially the dog, it is important that the animals have reached
stabilized baseline levels for the biochemical/clinical parameters before they are
started on test. Stabilization for many parameters, especially enzymes, often does
not occur before 4 months of age in dogs because of variability in age of maturation
and other reasons. For practical reasons and in keeping with the desirability of
utilizing young adults, starting dogs on test at 4 to 10 months is preferable to using
older animals.
Generally, young albino rabbits are used for dermal toxicity tests. It is suggested
that rabbits weigh between 2 and 3 kg at the start of the study. Selection of other
species or other age or weight ranges for dermal toxicity tests may be acceptable but
should be justified.
4. EVALUATION OF TOXIC EFFECTS
4.1 Introduction
A series of interrelated tests are suggested for the general evaluation of biological
systems, within practical constraints. In accordance with good laboratory practices,
physical and clinical laboratory procedures will be used as indicators of toxicity, and
tissue histopathology examinations will be performed on selected target organs.
Additional clinical, laboratory, and pathological tests may be employed if the indicator
evaluations suggest abnormalities. The underlying purpose is to use cost-effective
techniques to identify probable biological injury, which may then be examined in more
detail. The following section discusses these tests and suggests a test design that
provides a comprehensive evaluation.
The section is not meant to imply that only studies utilizing all these specific tests
or combinations of these tests will be accepted by the EPA for TSCA regulation. The
following discussions are included as examples of properly designed studies in which a
complete array of toxicity evaluations are interwoven to achieve a satisfactory level of
hazard assessment in a cost-effective manner. By indicating the perceived advantages
of each test, the Clinical Evaluations and Pathology Committee hopes to provide a
scientific basis for the selection of tests to include in a subchronic study design and yet
avoid becoming rigid and overly complicated.
-------
23
4.2 Clincial Observations and Evaluations
General clinical observations of experimental animals can include the periodic
(preferably daily) evaluation of general health, physical activity, and behavior, in
addition to observations on the function of specific systems, such as the
cardiovascular, reproductive, eye and special senses, and central nervous system. In
some instances, because of the animal species involved and the physical restraints of
size and anatomy, certain suggested examinations and procedures will not be
physically possible or practical (e.g., some procedures that are used in the dog may
not be practical in the rat).
4.2.1 Physical
Experimental animals should be observed daily by trained and experienced
laboratory and/or veterinary personnel (Balazs 1970; Benitz 1970; Janku and Krsiak
1973; National Academy of Sciences 1975; Prieur et al. 1973; Spurling 1977; Zbinden
1963). Physical observations include those of respiratory activity; appearance of the
eye (e.g., cloudiness of cornea, conjunctivitis, discharge); nasal discharge; mucous
membranes; skin color; and hair or coat appearance.
When practical, the heart and respiratory rates may be estimated by observation
or by palpation upon routine handling (Benitz 1970). Quantitation of these rates may
be performed when physical observations suggest a possibility of abnormality. Body
temperatures may be taken if the need is indicated by the results of other
observations, but in general, daily temperatures or exact pulse and respiration rates
are too variable to be very useful in toxicological testing.
4.2.2 Health
Evaluation of the following parameters should be performed daily on a qualitative
basis (Barnes and Denz 1954; Benitz 1970; Stevens 1977): appetite, presence and
character of feces, water consumption when practical, presence of urine, and
abnormal urinary consistency or color. When abnormalities are noted, urinary
volume may be measured. Each animal should be weighed and its food consumption
estimated at least once a week (Arnold etal. 1977; Barnes and Denz 1954; Benitz 1970;
Smith 1950). A weekly determination of weight is suggested because body weight is
often the only or earliest indication of toxic effects, and it ensures that every animal will
have a weekly physical examination.
4.2.3 Cardiovascular
A specific daily or weekly evaluation of the cardiovascular system is generally not
necessary, unless physical observations suggest abnormalities or unless chemicals
suspected of affecting the cardiovascular system are involved. Then special
examinations (e.g., electrocardiogram or blood pressure) may be performed (Benitz
1970).
Occasionally special monitoring of heart function will be an important part of
toxicological studies, because at a specific concentration some chemicals will affect
myocardial membranes enough to produce electrocardiographic changes. Among
-------
24
laboratory animals the techniques of collecting and evaluating an electrocardiogram
(EKG) is adequately developed only in the dog (Detweiler, Huben, and Patterson
1960; Detweiler and Patterson 1965). In addition to before dosing and at the end of the
study, EKGs should be taken during the first week of the study because significant but
transitory cardiotoxic effects may occur early in the dosing regimen. The frequency of
further testing is determined by the necessity to detect any suspected change early
and by the desirability of following the development of such cardiac alterations
including any possible recovery. The time of day the EKG is taken should be
determined in relationship to the metabolism and kinetics of the compound; generally,
it should be taken at the time of the chemical's peak cardiac effect.
4.2.4 Reproductive
The most reliable early detector of an abnormality of the male reproductive
system is histologic examination of the testes and epididymides, after proper tissue
collection and fixation. When the prior history of the chemical suggests a potential
effect on reproductive function, the male reproductive system may be monitored by a
semen evaluation obtained by electroejaculation or another method. Semen samples
may be evaluated for quantity and sperm motility, count, and morphology. This
evaluation is not a routine procedure.
During clinical examinations of male dogs, palpation of the testicles for abnormal
swelling and softness as well as the prominance and delineation of the epididymides
may be useful in detecting toxic insult to these organs. Subchronic studies do not
include an adequate detector of reproductive ability in the female; such detection
would require a separate specific study.
4.2.5 Ocular
On a routine basis the sight of the animal may be evaluated by observing its usual
movements within its housing area. Visual examination of the external eye is usually
performed as part of the routine physical examination (Balazs 1970). If visual defects
are suspected, further perception studies or more detailed ophthalmologic
examinations should be performed. Biomicroscopic (slit lamp) examinations may be
used for large animals (i.e., dogs and monkeys).
Ophthalmologic examinations are an essential part of subchronic toxicity studies
(Benitz 1970; Grant 1974; Meier-Ruge 1977). They should be conducted on a
representative number of the animals in the control and high-dose groups before the
study and at its termination, because ocular lesions generally develop slowly and are
seldom reversible (Grant 1974; Meier-Ruge 1977). If periodic or progressive changes
are expected and if it is necessary to document the development of such changes,
more frequent examinations may be performed. These examinations should be
performed by a qualified person.
4.2.6 Behavioral
Evaluation of behavior has traditionally been incorporated into toxicology
procedures (Balazs 1970; Benitz 1970; Janku and Krsiak 1973; Prieur et al. 1973;
Zbinden 1963). Many of the observations can be done routinely during the physical
-------
25
examination, which can include such tests as evaluation of the animal's attitudes (is it
alert or lethargic?), postural activities (movement and carriage of portions of the body,
limbs, and head), responsiveness to demand (ability to move when prodded), and
general behavior (is it suddenly aggressive or depressed from previous observation?).
Such observations may be easily performed by trained persons familiar with the
animal's usual behavior.
Evaluation of hearing may be determined by the animal's carriage and movement
of ears or by the introduction of unexpected external noise into the environment.
Such evaluations are relatively arbitrary and have not been standardized.
4.2.7 Neurological
Specific neurological examination may be performed monthly or when indicated.
In rodents, this evaluation may consist of balance beam or rotating-rod procedures; in
dogs more detailed studies may be performed, such as electroencephalograms
(EEGs) (DeLahunta 1977; Hoerlein 1971).
A detailed clinical and, when possible, instrumental neurological examination
should be made if it becomes apparent that treatment-related signs of nervous system
abnormalities are present. The objective of the test should be to determine the
location of the lesions so that ultimately these areas can be evaluated by
histopathologic or ultrastructural examination.
When neurological examinations suggest some abnormality, it is likely that
continued exposure to the chemical compound in question could produce
morphologic changes in the peripheral nerves or central nervous system. It may be
appropriate in such cases to proceed with special additional studies at higher doses
and more concentrated exposures to provide definitive morphologic-pathologic
changes for better interpretation of these toxicologic effects.
4.2.8 Memory and learning
Memory and learning studies are recent additions to special toxicologic studies
and are generally not included in routine subchronic studies. Memory and learning
patterns of rodents and dogs may be tested and evaluated during and following
exposure to chemicals. The procedures are currently limited to simplistic studies of
primitive learning patterns or abilities to remember trained or conditioned activities.
Mazes or pedal-pressing and reward trials may provide preliminary evidence of
alterations in nervous system function.
4.3 Clinical Laboratory Tests
The laboratory procedures used for monitoring animals in subchronic toxicity
experiments are usually tests that have been developed for diagnostic purposes.
Tests developed for one species are usually appropriate for another but need to be
modified because of quantitative differences. In monitoring animals in toxicological
studies to detect early deviations from normal, absolute precision, accuracy, and
sensitivity of test procedures are required.
The test selected should be appropriate for the animal species being utilized, be
selected to monitor a particular major organ or to react to a broad range of biological
-------
26
changes, and be the most sensitive in the normal range for that species. The test
procedure should be sufficiently precise so as not to add to the total statistical
variation of the results and allow comparison of results between laboratories.
4.3.1 Sampling frequency
The following suggestions are based on the use of the rat and the beagle dog as
the test animals.
Because of the undesirable effects caused by frequent blood sampling, care
should be taken not to collect at one time more than 10% of the animal's total blood
volume. More than this will stress the capacity of the animal to regenerate lost red cells
and would deplete the stored iron so that the data generated under these conditions
would be compromised.
If possible all animals inthe control and experimental groups should be sampled
at least once before the dosing is initiated. Half of the rats in each group should then be
sampled at one week, one month (and if the study warrants, every six to eight weeks
thereafter), and at the termination of the study. Precautions should be instituted so
that excessive sample removal does not affect body weight gain (Cardy and Warner
1979). The remaining half of each rat group should be sampled only at the study's
termination. Because of larger blood volumes, all the dogs in each group should be
sampled at each sampling period and termination. Because of the expenditure of
resources and the energies devoted in these types of studies, prior consultation with
biostatisticians is recommended.
4.3.2 Hematologic evaluations
General hemalologic examination consists of cellular evaluation and hemostatic
function evaluation. This might include packed red cell (PCV), which is the most
reliable method of all the erythroid measurements on peripheral blood (Linman 1966);
hemoglobin concentration (Hb), which is an excellent screening test (Wintrobe et al.
1974), especially the cyanmethemoglobin method recommended by the International
Committee for Standardization in Hematology (Lewis 1967); erythrocyte enumera-
tion (RBC), which is necessary for assessment of erythrocyte numbers and may be
required for the calculation of mean corpuscular volume (MCV); mean corpuscular
hemoglobin concentration (MCHC), which can be derived either from manual or
automated methods, which is remarkably constant throughout vertebrates, rarely
exceeding 37 g/dl (Medway and Geraci 1964; Hawkey 1977), and which can also serve
as a quality control measure verifying the compatibility ofHb and PCV; MCV, which is
directly measured in some electronic particle counters or may be obtained from RBC
and PCV values; total and differential white blood cell counts (WBC), including
absolute values, which are basic parameters for leukocyte evaluation; blood smear
examination including morphologic evaluation of erythrocytes, leukocytes, and
platelets and of their relationship to the quantitative data (Archer 1977; Wintrobe et al.
1974; Carrwright 1968; Linman 1966); prothrombin time (PT) (in dog only), which is a
nonspecific screening test measuring an overall functional ability of the extrinsic
coagulation system; and activated partial thromboplastin time (APTT) (in dog only),
which is a nonspecific screening test measuring an overall functional ability of the
-------
27
intrinsic coagulation system. Mean corpuscular hemoglobin is not suggested because
its usefulness may not be commensurate with the expenses involved in generating and
storing data.
Special tests, in addition to the standard screening tests, may be done when
quantitative parameters are different statistically from those of control values by ±2
standard deviations (or by other justifiable standards), or when qualitative assays or
clinical signs suggest a treatment-related hematologic abnormality. Special tests may
include but are not limited to reticulocyte count, platelet count (smear confirmation is
always necessary),Heinz bodies detection, methemoglobin determination, glucose-6-
phosphate dehydrogenase (G-6-P-D) estimation, pyruvate kinase (PK), platelet
aggregation, erythrocyte fragility, and iron determination and stain.
At termination a routine hematologic study is to be performed. If a treatment-
related hematotoxicity is present, a more complete hematologic evaluation including
PT, APTT, and bone marrow smear should be performed, together with
histopathologic examinations of the hematopoietic organs.
4.3.3 Clinical chemistry evaluation
The following is a list of commonly used clinical chemistry determinations which,
used as a unit, can provide a fairly complete chemical analysis screen of toxicity
(Kaneko and Cornelius 1970). However, it is not implied that these tests can totally
substitute for the use of histopathology or that these tests are the only clinical studies
that will be acceptable by the EPA. The intent of this discussion is to suggest a
practical clinical chemistry screen that may be used with histopathology to evaluate
the majority of the body systems.
• Glucose (fasting blood): Is the best single parameter of carbonhydrate metabolic
status and pancreatic islet cell intergrity. It also reflects nutritional state and
physiological stress. For example, it can reflect adrenal, thyroid, and, indirectly,
pituitary function.
• Glutamic-oxaloacetic transaminase (GOT): Reflects injury to the skeletal muscle,
cardiac muscle, and liver. Because it is found in high concentration in the RBC, it is
also useful as an indicator of the quality of the blood sample.
• Glutamic-pyruvate transaminase (GPT): Reflects liver injury in carnivores and
omnivores.
• Cholesterol: Reflects nutritional status, pancreatic islet cell damage (diabetes),
thyroid function, and cholestatic liver injury. In this unit of 13 recommended tests, it
is the best for reflecting thyroid function.
• Total protein: Reflects nutritional status, liver injury, intestinal injury, renal injury,
and several other pathological states.
• Albumin: Reflects protein imbalances and is required for proper screening
interpretation of the total protein.
• Globulin: Reflects the presence of some lymphoreticular tumors, acute inflamma-
tion, liver injury, chronic suppurative disease, renal disease, and several others.
Globulin is obtained by difference between total protein and albumin. If the protein,
albumin, or globulin falls outside two standard deviations from the mean or if other
-------
28
statistical parameters indicate chemical related variations, the proteins should be
electrophoresed to obtain accurate values for levels outside the normal range.
• Sodium: Permits evaluation of acid/base imbalances, adrenal function, and
intestinal injury.
• Potassium: Permits evaluation of acid/base imbalances, adrenal function, and
intestinal injury, but always used in conjunction with sodium.
• Alkaline phosphatase: May reflect bone injury, hepatocellular disease, and adrenal
hyperfunction (dog).
• Calcium: Permits assessment of the status of bone and of nutrition.
• Phosphorus: Reflects bone injury, renal disease, and nutritional status.
• Blood urea nitrogen (BUN): Reflects renal injury, prerenal and postrenal azotemia
and advanced liver disease.
All of the above tests can be performed with approximately 0.75 ml of serum,
which in the normal animal may be obtained from 2.0 to 2.5 ml of whole blood.
Because of the volume, the use of rats with body weight of less than 200 g is a limiting
factor. Therefore, it may not be possible to do all the suggested tests at all the sampling
intervals. However, all tests can be done on the blood collected at the termination of
the study. The combination of tests to be performed should be justified by the
investigator.
4.4 Special Organ Evaluation
When data indicate that a particular organ or organs are affected, certain
additional tests may be suggested to specifically assess damage to that target organ.
The merits of these tests should be properly evaluated. Tests would normally be
similar for all species tested, such as the rat and the dog, if practical aspects make their
use feasible. Pulmonary function tests are not specifically listed, because, except for
inhalation toxicity studies, lung changes would not ordinarily occur; but where such
damage might be present, pathological examination of lung tissue is the most
appropriate means of determining its extent and significance. When extrapolating
data to man, it is especially important to recognize the unique differences expected
between various experimental animal species in their responses to chemicals as
reflected in their organ function tests. It is necessary to have trained and experienced
comparative pathologists and toxicologists available for such interpretation. The
following discussion contains some suggestions for such assessments.
• Adrenal. In addition to sodium, potassium, and alkaline phosphatase, special tests
to provide a definitive clinical laboratory evaluation of the adrenal gland are cortisol
and the adrenocorticotrophic hormone (ACTH) response test.
• Bone. In addition to calcium, phosphorus, and alkaline phosphatase, a special test
is parathormone. Other histopathologic approaches may be more cost-effective.
• Brain and nervous tissue. If clinical signs and results of other tests warrant,
creatine phosphokinase (CPK) and isoenzymes may be determined, in addition to a
study of cerebrospinal fluid (chemistry and cytology).
• Intestine. In addition to total protein, albumin, globulin, sodium, and potassium,
special tests may include serum protein electrophoresis; amylase (dog); absorption
-------
29
tests for fat, glucose, and xylose; serum chloride; total plasma carbon dioxide; and
qualitative or quantitative tests for fecal fat, protein, and carbohydrate.
• Kidney. In addition to BUN, total proteins, albumin, and globulin, special tests may
include creatinine; endogenous urea clearance; endogenous creatinine clearance;
urinalysis; and serum protein electrophoresis.
• Liver. In addition to GOT, GPT, albumin, globulin, total protein, alkaline
phosphatase (AP), and cholesterol, special tests may include serum protein
electrophoresis, direct and indirect bilirubin, gamma gultamyl transferase (GGT),
lactate dehydrogenase (LDH) and its isoenzymes, Bromsulphalein (BSP) retention,
serum bile acids, and biopsy.
• Muscle. In addition to GOT, special tests may include CPK and isoenzyme, LDH
and its isoenzymes, and biopsy.
• Heart. In addition to GOT, special tests may include CPK and and its isoenzymes
and LDH and its isoenzymes.
• Exocrine pancreas. In addition to total protein, albumin, globulin, sodium, and
potassium, special tests may include amylase and lipase; serum protein
electrophoresis; fecal trypsin; absorption tests for fat, glucose, and xylose; and
qualitative or quantitative tests for fecal fat, protein, or carbohydrate.
• Endocrine pancreas. In addition to glucose, sodium, and potassium, special tests
may include intravenous glucose tolerance tests, insulin response to glucose loads,
and urinalysis.
• Thyroid. Special tests may include triiodothyronine by radioimmunoassay (T3),
thyroxine by radioimmunoassay (Tn), and the thyrotropin response test.
4.5 Urinalysis Evaluation
Although most toxicology guidelines recommend urinalysis as a test parameter
(Federal Register 1978; Food Safety Council 1978; Prieur et al. 1973; U.S.
Environmental Protection Agency 1979), it was the consensus that routine and
detailed evaluation of randomly selected urine samples normally has a minimal
usefulness in subchronic studies. If any urinalysis is performed, it is suggested that
nonrodents be used and rodents only when there is a need based on expected or
observed toxicity. Among the parameters that could be assessed are color, specific
gravity or osmolarity, pH, protein, glucose, ketones,formed elements (RBCs, WBCs,
epithelial cells, etc.), crystalline and amorphous materials, and blood pigments.
4.6 Pathology
4.6.1 Gross examination
Qualified scientists should perform or personally supervise the necropsies,
and/or be immediately available for consultation at the site. Other trained employees
may assist in the necropsies.
Animals must be necropsied as soon as possible after death. If necropsy cannot
be performed immediately after the animal is killed, or found dead, a trained
technician should immediately refrigerate (but not freeze) the animal at temperatures
low enough to minimize tissue autolysis (4-8°C). Animals found dead upon routine
-------
30
clinical examination should be necropsied as soon as possible to salvage usable
tissues. Animals found moribund should be sacrificed to assure the collection of
usable tissues.
The gross examination should include an initial physical examination of the
external surfaces and all orifices, followed by an examination of tissues and internal
organs in situ. The examination should include the following: external and internal
surfaces of all hollow organs; cranial cavity and external surfaces of the brain and
spinal cord; nasal cavity and paranasal sinuses; neck with its associated organs and
tissues; thoracic, abdominal, and pelvic cavities with their associated organs and
tissues; and the musculoskeletal system including testing manual-breaking strength of
a long bone or rib. (The latter procedure may be employed by the experienced
pathologist with reliable results and is more practical than elaborate instrumental
techniques.) In inhalation studies and where pulmonary damage is indicated, the lungs
should be inflated with a fixative to allow for better gross examination and
preservation. A minimum of 20% of the lungs from different groups should be so fixed,
with 50-75% fixed, in special instances, by the intratracheal perfusion technique.
The weights of the liver, kidneys, uterus, prostate (dogs only), and seminal
vesicles (except for dogs) should be recorded. In experiments involving chemicals
with suspected hypertensive effects, the weight of the heart should be recorded. Also,
in experiments in which the chemicals have a hormonal effect, the weights of the
gonads and the adrenals should be recorded.
4.6.2 Tissue preservation
The prosector should immediately preserve all tissues and organs from all test
animals in 10% buffered formalin or another recognized and accepted fixative
appropriate for the specific tissue(s) (Luna 1968). Sections (blocks) from the following
tissues from all test animals, regardless of their time of death, should be properly
preserved and routinely examined microscopically:
• all gross lesions, with a margin of normal tissues;
• brain, minimum of one transverse section each from the forebrain, midbrain, and
hindbrain;
• spinal cord, minimum of one section from the thoracic region;
• eye and attached optic nerve;
• a major salivary gland;
• thymus;
• thyroid;
• heart, two sections, one through a left papillary muscle and one through a coronary
artery;
• aorta;
• one lung with a major bronchus;
• stomach, fundic and pyloric regions;
• small intestine, three levels;
• large intestine, one level each of cecum and colon;
• adrenal gland;
• pancreas;
-------
31
liver, one lobe (a minimum of one left and one right lobe when gross lesions or other
evidence of hepatotoxicity are present);
gallbladder, if present;
spleen;
kidney;
urinary bladder;
mesenteric lymph node, plus tracheobronchial lymph nodes in inhalation studies
and regional lymph nodes in dermal toxicity studies;
bone, including marrow;
skin, only required in dermal toxicity studies and should be selected from skin
painting sites and normal areas;
• skeletal muscle, selected if there are clinical signs, gross lesions, or changes in
appropriate serum enzymes (contraction artifacts should be prevented and
enzyme histochemistry used when fiber types need to be identified); and
• reproductive organs, testes, and epididymides or uterus and ovaries.
The testes and epididymides, sectioned into 0.5-cm slices, should be preserved in
Bouin's solution or other appropriate fixative to ensure a high quality histologic
section. For detecting impairment of fertility in the male, this is the only method that is
routinely done in these end-point toxicity studies. Based on experience, or scientific
or practical judgments, other tissues may be preserved for later study.
4.6.3 Justification of selection of tissues for histopathological examinations
Histopathological examination of tissues is expensive, and in some organs it is
not the most reliable or sensitive detector of damage (Smith, Jones, and Hunt 1972).
Therefore, we have attempted to monitor changes in organs by those tests that are
most sensitive and least expensive. Microscopic examination of some tissues included
in other guidelines have been omitted from routine examination here when there is no
gross lesion, no change in appropriate clinicopathologial parameters, or no evidence
of toxicity from previous studies. However, monitoring of these organs has been done
by other means (e.g., weighing organs, clinical pathology, or necropsy observations),
although histopathologic examination may still be required in some studies. Except
where observations indicate, the histopathological studies should be conducted on
control and high-dose groups only.
For paired organs, it is necessary to examine only one if both are grossly normal
because exposure is via the bloodstream and thus change is assumed to be bilateral.
Only one lobe of the liver is sampled, if normal. If lesions are present, both a right and a
left lobe should be sampled. Since portal streaming occurs in some species, an orally
ingested toxicant could affect one lobe more than another (Popper and Schaffner
1974). For instance, toxicants absorbed from the small intestine could affect the right
lobes more severely.
Parathyroid function is usually monitored by checking the breaking strength of
bones at necropsy, the appropriate clinical pathology parameters (e.g., serum
calcium), and, in dogs, the size of the gland at necropsy. Because lesions in grossly
normal parathyroids are infrequently detected in Hematoxylin and Eosin-stained
(H&E) sections (Robbins 1967), histopathological examinations of the parathyroids
will not be done routinely.
-------
32
The trachea need not be examined microscopically in oral or dermal studies,
because they are unlikely to be exposed to significant concentrations of the test
material. If material is regurgitated and aspirated into the trachea and lungs, any
changes would be detected in the sections of bronchi and lungs. The trachea should
be examined at necropsy.
The esophagus is not routinely examined microscopically, because it has only
transitory contact with ingested material. It is examined at necropsy and sectioned if
there is a gross lesion. Irritant chemicals would produce gross lesions and/or clinical
signs.
The changes in weight of the uterus and the prostate-seminal vesicle are
monitored, because the weight changes are a more sensitive and earlier indicator than
microscopic examination. Early hyperplastic changes in prostates are especially
difficult to confirm by histopathological examination alone.
No separate collection of small arteries is required, as they will be examined in
numerous other tissues.
Routine histologic examination of apparently normal skin is unlikely to reveal a
lesion. Any skin lesion should be sectioned. The mammary glands can be examined
clinically and sectioned if any alterations in size or texture are detected during the
experiment or at necropsy.
Bone marrow function is monitored routinely by complete blood count (CBC).
Hematopoietic function is routinely monitored by CBC examinations and, when
necessary, bone marrow smears. At necropsy histologic examinations are made of
sections of bone marrow, spleen, lymph node, and thymus.
Histopathological examination of the H&E-stained sections of pituitary is not
likely to reveal treatment-related changes in a subchronic study. Where there is
clinical or other indication of pituitary dysfunction, serial sections and appropriate
special stains (e.g., modified trichrome stains) should be prepared (Luna 1968).
Routine examinations of the liver, gallbladder, kidney, urinary bladder, and
salivary gland are included, because these organs excrete foreign chemicals that are
absorbed into the body.
Treatment-related lesions in the heart are infrequent in experimental animals, but
because of the frequency of lesions in the human heart and aorta, these are routinely
sectioned. However, examination of these organs may not contribute significantly,
and this routine sectioning should be reevaluated when adequate data are available.
Visual function can be monitored through changes in animal behavior and by
ophthalmological examinations. However, one eye and its attached optic nerve will be
routinely cross- and longitudinally sectioned because of the possibility that animals
may have suffered visual impairment that escaped clinical detection.
4.6.4 Special considerations
Postmortem artifacts in muscle contraction may be prevented by clamping or by
allowing the muscle energy stores to be depleted before fixation (Adams, Denny-
Brown, and Pearson 1962; Price et al. 1965). If lesions are suspected, then skeletal
muscles, particularly the weight-bearing muscles of the pelvic and pectoral girdles
(Bradley 1978), should be examined at necropsy. During trimming, each muscle
should be identified and cut in longitudinal and transverse sections, and enzyme
-------
33
histochemistry (Dubowitz and Brooke 1973) and electronmicroscopy may be
necessary to characterize muscle lesions in more detail (e.g., to determine fiber types
involved).
In an inhalation study, multiple sections of the upper respiratory tract, external
nares, nasal turbinates, nasal cavity, paranasal sinuses, hypopharynxlarynx, and
tracheobronchial lymph nodes should be sectioned.
4.6.5 Central nervous system
Successful detection of lesions in the nervous system depends on examination of
the correct sites. Because examination of the nervous system is difficult and
frequently impossible, a thorough clinical examination may assist in characterizing the
alterations.
Where lesions of the central nervous system and peripheral nervous system are
suspected, transverse sections of the brain and cord are examined. Cord segments at
the levels of C2, C8, T7, and L& or Ly should be sampled. A minimum of six brain and
four cord segments is preferred for the initial examination of the central nervous
system. The nature and location of the lesions should be characterized.
For microscopic examination, brain and spinal cord sections should be chosen
based on clinical signs. In special cases, brain and cord may be fixed in situ by systemic
perfusion with 4% paraformaldehyde followed by 5% glutaraldehyde. In dogs with
clinical signs but nonlocalized lesions, the following transverse sections of brain
should be considered for selection: corpus striatum, optic chiasma and optic tracts,
mammillary bodies and hippocampus, rostral colliculus, caudal colliculus, cerebellum
and medulla at the level of the pons, cerebellum and medulla at the level of the facial
nerve nucleus, medulla at the level of the vestibular nuclei, and medulla at the obex.
Characterization of lesions should not depend on H&E-stained sections alone.
Selected electronmicroscopy, as well as special stains such as Holmes for normal
axons, Guillery for degenerating axons, Luxol fast blue for normal myelin, Glee's
modification of Marchi for degenerating myelin, osmium tetroxide staining of teased
peripheral nerves, and Holzer's stain for glial fibers may be used where appropriate.
4.7 Analysis and Presentation of Test Results
The report of the subchronic study should contain a summary and analysis of the
data and a statement of the conclusions drawn from the analysis. The summary
should highlight any and all lexicologically significant data or observations and give
the investigator's opinion as to the mechanism of the action of the toxicant, the target
organ or organs, and the sequence of pathological changes.
4.8 Usefulness and Correlation of Test Results
This section on the evaluation of toxic effects describes the usefulness and
correlation of sets of general observations, clinical laboratory tests, and pathology
examinations to be routinely evaluated in a subchronic toxicity study. It is suggested
that additional tests and examinations may be performed when appropriate. Table 4
shows how these three sets complement each other and allow evaluation of organs or
organ system(s) with the most appropriate and economical methods now available.
-------
34
Observations in subchronic tests that might trigger further testing were
considered. However, the variety of biological changes produced by chemicals is so
great and variable that no specific recommendations could be made.
Table 4. General observations, clinical laboratory tests, and pathology examinations
that may be used in subchronic toxicity studies
Organ or organ
system
Liver
Gastrointes-
tinal (GI)
system
Urinary system
Hematopoietic/
hemostatic
system
Nervous system
Eye
Respiratory
system
Endocrine
system
Reproductive
system
Skeletal
system
Cardiovascular
system
Skin
Muscle
General observations
Discoloration of mucous
membranes, edema, ascites
Diarrhea, vomit, stool.
appetite
Urine volume, consistency.
color
Discoloration of mucous
membranes, lethargy.
weakness
Posture, movements.
responses, behavior
Appearance, discharge,
ophthalmologic examination
Rate, coughing, nasal
discharge
Skin, hair coat, body
weight, urine and stool
characteristics
Appearance and palpation
of external reproductive
organs
Growth, deformation.
lameness
Rate and characteristic of
pulse, rhythm, edema,
ascites
Color, appearance, odor.
hair coat
Size, weakness, wasting,
decreased activity
Clinical laboratory
tests on blood
Glutamic oxaloacetic trans
aminase (GOT) glutamic
pyruvate transammase.
alkaline phosphatase (AP)
cholesterol, total pro
tem. albumm globulin
Total protein, albumin.
globulin sodium (Na)
potassium (K)
Blood urea nitrogen total
protein albumin, globulin
Packed red cell volume
hemoglobin, erythrocyte
count, total and diffe1'
ential leukocyte count
thrombocyte count, blood
srnear. prothrombin lime
activated partial thrombo
pldstin time
Total protein, albumin.
globulin
Glucose Na K AP (dog)
cholesterol
Calcium, phosphorous
AP
GOT
Total protein, albumin.
globulin
GOT. creatme
phosphokmase
Pathology examination0
L,verb
Stomach GI lr(ict qall
bladder (if prtspnt)
salivary gland
pancreas
Kidney and urinary
bladder6
Spleen t hymns mesen
tenc lymph nodes.
bone marrow smear and
section
Brain, spinal c. ord and
sciatic nerve
Eye and optic nerves
One lung with a major
bronchus
Thyroid adrenal
pancreas
Testes and epididyinis or
ovaries Uterus or
prosiate and seminal
vesicles6
Bone and breakage
strength
Heart,*' aorta small
arteries in other
tissues
Only in dermal studies
Only if indicated by
observations, clinical
chemistry or gross
lesiois
"All animals should undergo thorough gross examination, organs or tissues listed should be examined microscopically
6 These organs should also be weighed
------- An error occurred while trying to OCR this image.
-------
36
Table 5. Acute toxicity of nefopam in three species
Route LD5° L£>5°
(ma/kg)0 (95% confidence limits)
Mouse
Intravenous 45.5 36.5-54.3
Intramuscular 52.9 413-67.7
Oral 119 101-135
Rat
Intravenous
Intramuscular
Oral
Intravenous
Intramuscular
Oral
28
569
178
Dog
-20
-30
> 100 < 200
25.0-31.4
47 4-68.3
146-217
Mg per kg of body weight
Source: Adapted from Case, Smith, and Nelson, 1975.
the route of administration (Table 6). He further suggests that comparative LD50's of a
compound given by different routes may allow prediction of information concerning
translocation, deposition, inactivation, or site of excretion of the compound in
question. He does not, however, suggest a rationale for estimation of toxicity of a
compound given by one route when information is available about the toxicity when
given by another route. Limited data from studies not appearing in the available
literature suggest that route-to-route correlations and subsequent estimations are
possible for some compounds but not for others.
Table 7 compares the results of two 90-d toxicity studies in rats with vinylidene
chloride (VDC). In both studies the primary toxic sign, degenerative fatty liver
changes, observed at daily doses of 28 mg/kg in the drinking water study and 42 mg/kg
in the inhalation study, was described as being slightly more severe by the inhalation
route. The correlation was also observed when similar experiments were performed
on fasted and nonfasted rats, even though toxicity was more severe in fasted than in
nonfasted rats. It appears that with this approach, which uses breathing rates of rats
and chamber concentrations and assumes 100% absorption, there is a positive
correlation between the daily dose on a mg/kg basis and the observed toxicity for two
routes of exposure.
Table 8 compares the toxicity of perchloroethylene (PERC) administered in rats
by either gavage or inhalation. In the gavage study PERC was administered in a corn
oil solution daily at a dose of 500 or 1000 mg/kg for two years. In the inhalation study
rats were exposed to 10 or 600 ppm of PERC for one year followed by one year of
observation without exposure. Both studies reported that the only toxicity observed
was the early onset of mortality in males due to chronic renal disease. (Males of this
strain of rat have a naturally high rate of spontaneous chronic renal disease.)
Calculation indicates that doses of 10 and 600 ppm for 6 h/d approximate daily doses
-------
37
-
~S c
t .2
&|
3 '£
.s|
C "
o
DC ^S
0)
Q
SCO
y
3 ^
~—-
O ' — ~
15 •>
DC v2L
^
^3
to
cc
in Ji!
Q "Si
-J E
'~~^
O ^
c£ 3;
O)
U)
Q ~^5
-J £
o"^-v
4^ >
{£ >
QJ
§
S u
0 -^
Q ~O5
-J ^
C
0
11
3-1
& 1
"S
o vp LO vo LH
,__,_,_
o o ^t o o
co ro csj co co
1— 1 !— 1 I i CM
LO
CM
I
1-1 CM OJ CM ^
*^ O LO O Q^
O I-H O i— ' ^*
O O C^ O CJ>
I-H O O •-< O
CO CM O O CM
LO CO ^ *>O ^J"
•-I ^> ^ CO rH
iO O O O O
^t CO CO O O
CM ^> 00 LO
16 ^ w>
» s^ §
§ s 1 s
C s- T fO
I llj-a
1 1 J ^ 6
s ?
"E ^
13 s-
ro oj
yo cc
QJ
-u" I
QJ C
o ^5
o r^
0) U
Cu M-
c/) o
0) -K
C/)
"> i-1
•< QJ
. >
w c
•a D
G 03"
1 1
c
"5 "^
^ o
rn" "S
0 1
> D
^ 1
3 I
S 8
^ Q
-^ r>
o ^-«
"O O
S "^
•-C c
£ ^
^ w
TO f r\
Q ^
® ~c
"ra TO
CX QJ
(/) c
o b
a . CQ
O ^£)
o Q> Q
Ij E
O Q. O
Q, d) ^*-
21 SS
S|a -
a -° ^r
o "ro
U O
01
c
"5
n
S
T3
QJ
U)
0)
~I3
0
-X
c/)
O
C
0)
I-*
"E
QJ
j;
*^
13
C
~ &
-^S o
cy ro S,
^ t- ^r~
3 0 ^
0 -S
£ -"
JS -^ £
Jc "^ O S
CO ^ -£ _j
5 a ^ §
>. oj £ i_
Q -^ g ^
JD c 'J~ cp
V4~ ^ -2 "D,
° &» TO TO
en j| "D TJ
II ^ ^
2 K tx o
j -a a, CO
-------
Table 7. Comparative results of 90-d toxicity studies in rats administered
vinylidene chloride by drinking water or inhalation
Drinking water Inhalation"
Concentration (ppm)
Daily dose (mg/kg)Q
Mean dose
Degenerative
fatty liver
50
5-12
Q
^ O
Absent
100
8-20
-14
Absent
200
16-40
-28
Present
25
14b
Absent
75
42 c
Present d
Exposure for 6 h/d, 5 d/week.
Mg per kg of body weight.
cCalculated using breathing rate of rat and the concentration in air
d Slightly more pronounced than that observed in the drinking water study
Source: W. Braun, Dow Chemical Co (unpublished data)
Table 8. Comparative toxicity in rats administered perchloroethylene
by gavage or inhalation
Oral gavage Inhalation
Concentration (ppm)
Daily dose (mg/kg)°
Early mortality males
500
Absent
1000
Present
10
20
Absent
600
1200
Present
(chronic renal disease)
°Mg per kg of body weight
Source: Pegg et al., 1979
of 200 and 1200 mg/kg respectively. When this approach is used, there again appears
to be a correlation between the observed toxicity studies and the daily doses
expressed on a mg/kg basis.
These apparent correlations of dose with observed toxicity for two different
routes of exposure should not be taken as a general correlation that will hold in all
cases. The correlation noted for PERC is most probably reasonable and is easily
explained, because Pegg et al. (1979) have demonstrated that the pharmacokinetics
and metabolism of PERC are independent of the route of exposure. However, when
pharmacokinetics and metabolism are dependent upon route of exposure,
correlation with effect may not be apparent and extrapolation becomes more difficult.
Table 9 summarizes the results from two chronic tox icity studies in rats with 1,4-
dioxane administered either via inhalation or drinking water. Previously mentioned
dose calculations indicate no evidence of correlation between dose and observed
toxicity. Young, Braun, and Gehring (1978) demonstrated that blood concentrations
of 1, 4-dioxane were comparable for estimated equivalent doses by either drinking
water or inhalation exposure. Liver concentrations of 1, 4-dioxane, however, were
much higher after oral administration.
One of the factors that could form a potential basis for extrapolation, with some
degree of confidence, from one route to another would be an equivalence in peripheral
-------
39
Table 9. Summary of results from two chronic toxicity
studies in rats with 1,4-dioxane
Drinking water Inhalation
Concentration (ppm)
Daily dose (mg/kg)a
Toxicity
100
Liver
toxicity
111
1000 105
Tumors No toxicity
in liver or
othei organs
° Mg per kg of body weight
Source: Adapted from Young, Braun, and Gehnng, 1978.
blood concentrations. However, this principle would not apply in situations like the
following.
• The organ of entry is the target organ. If an inhaled compound produces an effect on
the lung, it would be unlikely (but not impossible, because there are systemic lung
toxicants) that the effect would be produced by ingestion. Similarly, if ingested
materials cause stomach damage it would be difficult to extrapolate that effect to a
different route.
• The organ of entry to another organ or system is without the benefit of detoxication.
Two examples are apparent. The effects of inhaled materials on the heart and the
effect of ingested materials directly on the liver. In these examples the heart and
liver are exposed to the material directly as it comes from the organ of entry. The
concentration in the pulmonary and hepatic veins may be far greater than that in the
peripheral blood, and thus the potential for effect is much greater for the heart and
liver.
• The compound is highly reactive. In this case administration by various routes may
result in totally different target organs or metabolites.
Generally, it is reasonable to assume that once a material enters the circulatory
system, its subsequent behavior is independent of route of administration. Thus, if the
blood concentration is known for two routes of administration as a function of time
and dose (exposure) and if the toxic effects via one route are known, prediction of
toxicity by the other route is possible. In the absence of either blood kinetic
information or exposure-blood level information, the prediction of toxicity by a
different route will be highly uncertain.
In summary, few studies allow for estimations of the type mentioned above, and
many studies indicate that extrapolation from one route to another would be
impossible. In the absence of any other information and where some estimations are
needed, the approach as outlined by Stokinger and Woodward (1958) could be
followed.
This approach used retrospective studies and standards for extrapolating data
from various routes of exposure to develop drinking water standards. Because much
work had been done on the classes of chemicals they were dealing with, the authors
relied heavily on established standards such as Threshold Limit Values for industrial
-------
40
chemicals in air, federal tolerances for pesticide residues, toxicological classifications
by type of compound, and toxicity data from nutritional and other studies.
The authors used similar adaptation techniques for data generated in animal tests
by various routes. The following summarizes the basics of the extrapolation process.
• For data from inhalation studies, LD50 or related values should be converted to a
common dosage unit (i.e., mg/kg or mg/kg per day) through the use of standard
respiratory parameters for the subject species over the exposure period. This
adaptation must include the development of absorption factors that estimate the
amount of test material absorbed into the blood stream. Use of these factors
introduces greater uncertainty into establishing uniform exposure situations, but
data from kinetic studies and physical characterization of the test material may
diminish the uncertainty somewhat.
• For ingestion data, toxic values should be expressed in standard units suitable for
conversion, such as those listed above. An absorption factor must again be
developed by using whatever data is available to diminish the inherent uncertainty.
• For dermal data, suitable dosage units can be calculated (but exposure times must
be carefully studied in light of the test design being reviewed).
• For parenteral data, similar standards can be developed, but absorption factors,
depending on the materials being studied, may be made with greater confidence.
After conversion of known data to uniform dosage units, it follows that the
reverse process is used to calculate exposure concentrations, applied doses, etc.
Thus, the only available reference that suggests a reasonable approach to the
problem of extrapolation from one route to another appears to be Stokinger and
Woodward. In essence, it suggests a considered judgment of all available toxicological
data in attempts to carry out such an extrapolation.
5.3 Factors Affecting Extrapolation of Data Obtained by One
Route of Exposure to Another
The effect of route of administration on absorption, distribution, metabolism,
excretion, and potential toxic response is well known. However, the committee felt
that a brief review of the subject was worth including.
"As a general rule... absorption of chemicals will be most rapid when given by
inhalation, less rapid when given by gavage, and slowest with dermal application"
(World Health Organization 1978, p. 68). The order is affected by the role of physico-
chemical factors in absorption. In general the skin is not highly permeable to most
chemicals and acts as a reasonable barrier against environmental chemicals. Systemic
injury and death can result, however, from some chemicals which pass through the
skin with relative ease. Klaasen (1975) points out that for most materials a large
number of dermal cell layers act as barriers to absorption through the intact skin in
comparison to absorption through the lung and gastrointestinal tract, which
represent only a one- or two-cell layer barrier. Because the intact stratum corneum
represents a major barrier, chemicals that injure this structure may be more rapidly
absorbed.
-------
41
Substances present in the lung or gastrointestinal tract have in common the fact
that they are essentially outside the body and produce only localized surface effects
until they are absorbed into the bloodstream. Chemicals absorbed from the lung are
carried directly to the heart and then into the systemic circulation. In contrast, most
compounds absorbed from the gastrointestinal tract are carried directly to the liver,
which in many instances may be a direct target organ. Gastrointestinal absorption, in
contrast to lung absorption, is complicated by several functions of the liver. Because
of its size and enzymatic activity, the liver is the major site of metabolism of most
xenobiotics and is often the target organ of the metabolites formed within its cells. A
variety of chemicals may be excreted through the bile either unchanged or in the form
of metabolites. Once returned to the gastrointestinal tract, these compounds may be
reabsorbed to produce an additional toxic impact on the liver.
In summary, the initial impact of chemicals absorbed from the lung is
considerably different from effects following absorption from the gastrointestinal
tract. Thus the portal of entry may play an important role in the toxic response,
particularly where the immediate organ involved (lung, liver, or skin) is the target
organ. Target organs different from the portal of entry and exposed strictly from the
systemic circulation would be expected to respond directly to comparable blood
levels of compound or metabolites regardless of the route of entry.
Physico-chemical properties of the chemical may influence the extent of systemic
absorption by various routes of exposure and thereby the development of the test
procedures. This is discussed in earlier sections.
5.4 Assumptions and Criteria for Extrapolation
The following assumptions were discussed by the committee.
• Given quantitative and temporal equivalency of circulating blood levels of a
chemical, it is reasonable to assume that systemic effects will be comparable
regardless of the route of administration.
• In making route-to-route extrapolations, only systemic effects other than those
occurring at the portal of entry should be considered or compared.
• Effects at the portal of entry should be considered and evaluated separately.
When results of a subchronic toxicity test are to be considered as definitive, the
committee recommends that a study by the primary human exposure route be
conducted. Under the following conditions, study by the primary route is totally
impractical.
• In certain cases the vapor pressure of a compound is low, yet exposure is to the
vapor. It would be difficult to produce vapor concentrations high enough to induce
toxicity in animals. Particulate aerosols could be produced, but exposure to these
would not adequately represent exposure to the vapor. In this case exposure by
ingestion might be entirely appropriate.
• Certain industrial dusts are produced in a unique manner (e.g., fiber coating dust in
making fibers). In this case, ingestion may be the only feasible route of
administration.
-------
42
• Vapors and organic dusts are possible explosives. While every effort should be
made to conduct vapor exposures via inhalation, an inhalation study may not
always be possible. The hazards of dust aerosols could preclude inhalation studies.
• Cost and availability of compounds may at times preclude inhalation studies,
particularly those involving whole body exposure, which are very wasteful of test
materials, and those using developmental products.
The Route-to-Route Extrapolation Committee considered the extrapolation of
subchronic data obtained by ingestion of chemicals in the diet to predict effects of
inhalation exposure to be the most likely to be employed and the most scientifically
acceptable extrapolation. Inhalation-to-ingestion extrapolations are probably feasible
and scientifically reasonable, but will be rarely needed. Extrapolation and estimation
of systemic effects via the inhalation or oral route from data obtained in dermal studies
was considered to be feasible but not very desirable because of a lack of defined
methodology for conducting dermal subchronic studies. Extrapolation to the dermal
route was not considered to be a probable need.
The validity of any extrapolation should be evaluated when toxic signs developed
in testing by one route indicate that changes might occur in parameters that affect the
absorption, distribution, metabolism, etc., by the untested route.
The committee recognized the need for separate consideration of extrapolation
criteria to be applied in situations where data are in existence (retrospective analysis
and extrapolation) and where data are to be developed (prospective analysis and
extrapolation).
5.5 Retrospective Analysis and Extrapolation
Retrospective analysis involves the gathering and evaluation of all available data
on the compound of interest. The objectives of the analysis are (a) to render a
judgment as to whether the available data are adequate for making an extrapolation
without developing additional data and (b) to determine what additional data are
needed if existing data are inadequate. In retrospective analyses, the existing data are
likely to range from no data to data from a considerable number of studies that relate
to numerous toxicological responses.
The following are suggested as guidelines for data evaluation.
• The subchronic study (or studies) from which data are to be extrapolated must
have been adequately designed and conducted to allow for proper application of
extrapolated results. For example, a subchronic study designed and conducted to
evaluate compound-related effects in only one tissue or organ would not be
considered adequate for risk assessment purposes.
• Data from studies on species or animal strains different from those used in the
subchronic study are not acceptable for extrapolation purposes.
• At least one study that assesses compound-related effects at the portal or organ of
entry by the route to which extrapolation is to be made is strongly recommended.
• At least one study of the same type and duration (e.g., LD50, LC50, EC50, etc.) and
utilizing the same end point(s) by each of the two applicable routes of administration
is essential.
-------
43
• Studies that make route-to-route comparisons of such things as pharmacokinetics
and metabolism are highly desirable. At a minimum, comparative blood level data
should be available if any degree of confidence is attached to extrapolated results.
Data on the other parameters can substantially increase confidence in extrapolated
results.
• Comparative studies that assess biochemical, hematological, clinical, pathological,
or other parameters can also increase the degree of confidence.
Under conditions when it would not be practical to perform a study by the
untested route, the procedure presented for prospective analyses is recommended.
5.6 Prospective Analysis and Extrapolation
Prospective analysis, as envisioned by the committee, is limited to those
situations in which exposure by the primary route of anticipated human exposure is
impractical and in which either no data exist or existing data are inadequate for
extrapolation purposes.
In these situations, the investigator has the advantage of being able to plan
carefully all studies involving extrapolation.
For prospective extrapolations, the committee considers the following as being
representative of ideal requirements:
• data that demonstrate a similarity in acute and/or subchronic response(s)
(including lethality) to the test material by both routes of administration;
• data that demonstrate equivalency in dose as evidenced by blood levels, total body
burden, or similar parameters;
• data that reasonably support the conclusion or assumption that pharmacokinetics
are independent of route of administration;
• data that reasonably support the conclusion that effects at the organ or portal of
entry by the tested route would or would not be anticipated to occur as a result of
exposure by the untested route; and
• data that address the possibility of effects at the portal of entry for the untested
route.
Given the knowledge that development of all ideal data will not be feasible, the
committee considers certain combinations of these data to be most desirable.
The best situation for extrapolation would permit a 14-d study by both routes of
administration. Such a study would serve to establish the target organ(s) and to
demonstrate direct effects on the organ of entry. If the same toxicologic end point
were observed after dosing by both routes, extrapolation from one route to another
after a 90-d study could be carried out with reasonable confidence. In the absence of a
common effect, the concentration of the test material in blood during the 14-d studies
could be used to predict results from an alternative route. The error associated with
this extrapolation would be somewhat greater than in the previous case.
If 14-d comparisons cannot be performed, comparisons of LD50 data between
each of the routes could be generated. This would serve to equate the doses by the
two routes. However, because the cause of death after acute studies may not be
related to subchronic toxicity, this approach would be less desirable.
-------
44
If data comparing toxicity as a function of route cannot be obtained, the
extrapolation is more difficult. However, when a need exists, the following information
would be useful:
• physical and chemical properties (solubility in water, acids, saline, octanol-water
distribution coefficient, particle size in the atmosphere, and pH);
• pulmonary toxicity by intratracheal routes;
• any information on distribution, metabolism, retention, and excretion;
• target organs; and
• in vitro studies.
5.7 Summary and Conclusions
• any extrapolation of toxic dosage levels from one route to another must be
considered in the light of the associated level of uncertainty. Situations exist
wherein valid extrapolations cannot be made.
• The justification for extrapolating data rather than performing the subchronic test
should be based on scientific rationale rather than on economics alone.
• The literature base for making valid extrapolations is almost nonexistent, although
much data may exist either in unpublished form or in a form not readily identifiable
through normal search procedures.
• Most extrapolations would be made using data from subchronic oral dosing to
predict inhalation effects, although the reverse may also occur. Some extrapola-
tions may be made using acute and subchronic dermal studies to predict toxicity by
other routes, but the reverse is not practical.
• All available sources of information should be sought out to minimize the
uncertainty of such extrapolations. Data from acute, short term, and subchronic
studies from all available routes, human exposure, and kinetic and physical-
chemical input will be needed.
• Extrapolations of this type should be performed by qualified scientists.
6. LIMITATIONS OF ACUTE AND SUBCHRONIC TOXICITY STUDIES
6.1 Introduction
The rationale for the use of the results of toxicity studies in animals and suggested
limitations of these studies are summarized in the following paragraph taken from the
World Health Organization (1978) document:
In many cases, studies with laboratory animals make it possible to predict the
toxic effects of chemicals in man. However, it is important to realize that experimental
animal models have their limitations, and that the accuracy and reliability of a quanti-
tative prediction of toxicity in man depend on a number of conditions, such as choice
of animal species, design of the experiments, and methods of extrapolation of animal
data to man. (p. 35)
Because there is always some uncertainty in extrapolating the results of toxicity
studies in animals to man, a safety factor is introduced to compensate for this
-------
45
uncertainty. The World Health Organization document addressed this point as
follows:
In general, the size of the safety factor will depend on (a) the nature of the toxic
effects, (b) the size and type of population to be protected, and (c) the quality (and
quantity) of toxicological information available, (p. 38)
6.2 Acute Toxicity
Acute toxicity studies (oral, intravenous, dermal, ocular, inhalation, etc.)
demonstrate biological potency. On this basis an estimate of hazard may be predictive
for a single, accidental human exposure. This extrapolation is not always valid, but
interpretation of the results may be strengthened by comparison with results for
similar chemicals.
6.3 Interpretation and Limitations of Acute Toxicity Studies
The single-dose oral and parenteral (intraperitoneal, intravenous, inhalation,
ocular and/or dermal) toxicity studies provide a relatively good estimate of the
potential local or systemic toxicity of a single dose to man. The results can help define
the severity and range of toxicity according to dose, species, and sex and may suggest
the need for subchronic tests. The results should include descriptions of the clinical
signs of toxic effects at all dose levels in addition to the determination of the LD50. The
slope of the dose-response curves can be used to estimate safe levels of exposure for
man. A steep dose curve indicates a relatively narrow range of effect doses, and
therefore the estimation of the safe exposure level will be more accurate than if there is
a shallow dose-response curve. The shallow dose-response curve, particularly if
toxicity is delayed, suggests a complex biological action of the chemical and therefore
a more variable individual response to the chemical (Casarett 1975).
The results of these tests give an indication of the clinical signs of toxicity
produced by the material and may help to determine the potential risk to humans
following oral, dermal, ocular, or inhalation exposure. There are limitations of these
tests that need to be considered when extrapolating to human exposure conditions.
• There are many species-sex variables that make interpretation of the tests difficult.
• Variable results can be produced depending on the vehicles used in the tests.
• High doses may alter normal metabolic pathways because of overloading of
metabolic enzymes and/or exhaustion of substrates required for conjugation.
• Results from single-dose studies are not reliable for predicting the hazards
associated with continuous low-dose exposure.
• Results from one route of administration may be difficult to extrapolate to other
routes.
• Tests should still detect significant potential hazard, although inter- and
intralaboratory variation may occur.
From acute toxicity studies there are no universal triggers that can be used as
guides to require subchronic studies. The determination of the need for subchronic
studies should involve not only the magnitude of the dose but also the results of close
observation of the animals. Delayed deaths may be the result of biochemical,
-------
46
hematologic, or anatomical changes or of persistence of the chemical in the body.
However, these changes may be due to the excessive doses used to produce lethal
effects. If signs of toxicity persist at the end of the observation period of 7 or 14 d,
particularly among animals at the lower-dose levels, further investigation, including
subchronic studies, may be necessary.
6.4 Subchronic Toxicity
Subchronic (90-d) toxicity studies provide a good estimate of a chemical's
potential hazards. These studies reveal essentially the same toxicity pattern (with the
possible exception of ocular and oncogenic effects) as a chronic studies, although the
no-observable-effect dose will be greater than the chronic no-observable-effect dose.
This should be considered when evaluating the hazard and assessing the risk
associated with the use of the chemical.
Because extensive biochemical, hematological, and anatomical studies are
included in the subchronic experiment, certain reactions may trigger the need for
additional studies. Also, because the dose range used in the subchronic studies
includes both toxic and nontoxic doses, there inevitably will be changes in one or
more of the large number of laboratory tests used as well as in the pathology studies.
These changes must be evaluated carefully. Only if they occurred with increasing
severity from the low-dose level to the high-dose level should additional studies be
considered. Examples of such triggers are hyperglycemia, particularly with changes of
the beta cells of the pancreatic islets; alternation of the differential white cell count,
particularly if a neutropenia is evident; decreased testicular weight, particularly with a
decrease in the number of sperm; and unusual hyperplastic change beyond that which
could be expected as a physiological alteration.
Conventional subchronic studies are of limited value in defining potential adverse
effects on the areas of behavior, species variation, reproductive physiology, and
immunotoxicology. Furthermore, the conventional subchronic study is of virtually no
value in defining potentially adverse effects in the areas of teratogenesis, genetic
effects, longevity, and neoplasia.
6.5 Special Considerations
6.5.1 Reproduction and teratology
Changes in reproductive tissues as determined in subchronic studies may
suggest the need for fertility and/or reproduction studies. The selection of the
maximum dose for teratology studies is critical, because a maternal toxic dose may
indirectly cause fetal malformations. Therefore, teratology studies should be done
only after completion of subchronic studies.
6.5.2 Recovery
Recovery studies may be needed to determine if an adversely affected parameter
returns to normal after termination of exposure. The lack of recovery at low doses
may dictate the need for additional long-term studies usually at the same and lower
dose levels.
-------
47
6.5.3 Pharmacokinetic and metabolic data
One has to make a judgment about the need for pharmacokinetic and/or
metabolic data. These data can be used to facilitate or clarify human risk assessment
or to select the appropriate species for subchronic studies. In this instance, studies of
absorption, distribution, metabolism, and elimination (ADME) on pertinent species
with particular emphasis on biologic half-life may be appropriate. In specific cases,
pharmacokinetic studies, including the identification of important metabolites, may be
appropriate. The need for metabolic studies should be determined by sound scientific
judgment based on the expected exposure of humans and the degree and extent of
toxicity observed in animals.
It is clear that there are scientific and economic bases for deciding what species
will be used. From a scientific reference point, one would recommend the use of that
species for which the ADME is most like that of the human (if this can be deduced).
The economic bases for selecting species, cost and availability, may influence the
decision.
6.6 Dermal and Ocular Studies
In the industrial situation, unless adequate protection is provided, the most likely
exposure to chemicals (excluding inhalation) is by dermal and/or ocular exposure.
Dermal exposure is particularly important because of the potential for systemic
exposure by penetration of the chemical through the skin. Although the rabbit is the
most frequently used species, its skin is more permeable than human skin for most
compounds (Bartek, La Budde, and Maibach 1972). The animals in which
characteristics of cutaneous penetration most resemble those in the human are the
rhesus monkey and the domestic swine (Wester and Maibach 1977). Other aspects of
potential toxicity due to dermal exposure are irritation (Mathias and Maibach 1978),
contact sensitization (Magnusson and Kligman 1977; Marzulli and Maibach 1977),
phototoxicity (Maibach and Marzulli 1977), photoallergy (Harber and Shalita 1977),
hypopigmentation or hyperpigmentation (Gellin, Ring, and Maibach 1977), and
chloracne (National Academy of Sciences 1977). In these various tests, extrapolation
to the human requires consideration of species, site of application, dose, vehicle, and
method of application. Some of the limitations of dermal studies that affect
extrapolations are as follows:
• Rabbit skin is more permeable than human skin for most compounds.
• Manyfold differences in permeability occur from the least to the most permeable
anatomic sites.
• Nonperspiring rodent or rabbit skin does not directly mimic human skin.
• Sufficient information is often unavailable for extrapolating from the mouse, guinea
pig, hamster, or dog to the human.
• Use of occlusive dressings on test animals exaggerates human exposure.
• Duration of exposure (i.e., 24 h) in test animals exaggerates human exposure (i.e., 8
h/d).
Eye irritation tests (Draize 1959) give an index of eye irritation potential in the
absence of human exposure. These tests should be used for comparison with groups
-------
48
of closely related compounds for which human data are available. In the rabbit test a
positive result should be assessed along with other factors affecting human exposure
such as conditions of use and misuse. When the test results in the rabbit suggest
potential hazard, tests in the rhesus monkey may be useful as an aid to evaluate
hazard for humans (Buehler and Newman 1964; Beckley, Russel, and Rubin 1969).
6.7 Conclusions
Valuable information for assessing the risk of a chemical can be obtained from
acute and subchronic studies in animals. The doses used in such studies (nontoxic to
toxic levels) and the duration of the subchronic study (about 90 d), with a wide variety
of laboratory tests and ultimately pathology, provide a good evaluation of the
toxicologic attributes of that chemical and may also determine whether additional
studies are necessary. Almost invariably acute systemic toxicity studies would also be
available if a subchronic study is to be done, because the need for a subchronic study
is usually indicated by the results of an acute study, which is also necessary to permit a
more scientific selection of dose levels for the subchronic study. Once the toxicologic
attributes of a chemical are known, certain risk evaluations can be made on the basis
of the type of toxicity and the possible exposure of humans. Acute toxicity studies by
themselves can be of some value, particularly for single exposure for the human, but
they are unreliable for more than a suggestive risk if continued exposure is probable.
The dermal and ocular acute tests again are useful for very short-term exposure that
most likely would occur in the use of the chemicals. However, single dermal exposure
would be of little, if any, value for the determination of systemic toxicity unless
penetration is so rapid that substantial blood levels are obtained with consequent
potential toxic effects.
7. UNRESOLVED ISSUES AND RESEARCH
RECOMMENDATIONS
The committees identified the issues and research needs associated with
subchronic testing. The issues identified are the ones that could not be resolved
because of a lack of literature and time. The committees were to recommend the
appropriate research solutions for these areas and issues. They are as follows:
7.1 Unresolved issues
1. The use of nonrodent test species: Available literature should be reviewed to
determine whether nonrodent species reveal unique or different signs of toxicity
with sufficient frequency to warrant their routine use. This would include published
information and data solicited by a general appeal to academic institutions,
governmental agencies, and industrial concerns.
2. Duration for dog studies: It is recommended that comparisons of the literature data
from dog studies in three-month and two-year durations can and should be made. If
the former adequately predict the latter, the dog should be included as a species in
the subchronic studies and not in nononcogenic chronic studies.
3. Number of animals: The exact number of test animals and the associated
appropriate statistical treatments should be evaluated.
-------
49
4. Relationship between chemical class and toxic effect: A data base needs to be
generated that will aid in the correlation of specific tissue effects with chemical
structure and activity. This will aid the prediction of toxic effects and assist the
investigator in the protocol design for specific chemicals. Also, it would be useful to
include similar information on chemical classes and organ systems.
5. Extrapolations from route to route (factors affecting the feasibility):
a. An in-depth review of the literature should be undertaken to find data where the
same chemical has been evaluated by different routes of exposures. This review
should include acute, subchronic, and chronic data.
b. A critical evaluation of such available data should be undertaken to establish for
specific materials, or classes of compounds, whether target organs and overall
animal response are comparable by the two routes.
c. If variable dose and quantitative response data are available, a reasonable basis
for extrapolation may be developed for individual materials and possibly groups
of chemical compounds.
7.2 Research Recommendations
1. Comparative experimental studies of 14-, 30-, and 90-d durations are needed to
evaluate the usefulness of the shorter period. These studies could include
comparisons of various arrangements of toxicity parameters (e.g., enzyme
function tests, hematology tests, pathology) that would help determine the
appropriate tests necessary for the different durations.
2. Research is needed to determine the applicability of pharmacokinetic methods to
the development of models for interspecies extrapolation. Differences in the
responses of species to specific chemicals can be related to differences of half-
lives of the parent compounds or toxic metabolites. In addition, the route of
elimination can vary between species with an increase of toxicity resulting in
specific organs of elimination in one species and not another. Pharmacokinetic
methodology could also reveal the overloading of enzyme systems that may be
deficient in one species as compared to another.
3. The similar metabolism of a compound in animal species suggests that their
biological responses to toxic chemicals may be similar. Therefore, a rapid in vitro
metabolic experiment using microsomal enzyme preparations from the liver or
other indicated organs from several animals, including man, may aid in the
selection of the most suitable animal for subchronic experiments. A data base
should be compiled that lists known differences in biological response between
species, including humans, and correlates these with species differences in
ADME.
4. To evaluate the potential use of oral studies to determine inhalation effects, acute,
two week, and subchronic 90-d studies should be performed by both routes.
Included in these studies should be pharmacokinetic data to ascertain the
development of steady state blood levels and tissue distribution. Histopathologi-
cal data should be obtained on both short-term and subchronic studies. These
data should be used to demonstrate the validity of extrapolation from one route
to another and to develop extrapolation factors.
5. Research is needed to determine the degree of species specificity for the
functional and biochemical tests. Because these tests were originally used for
-------
50
diagnosis of human diseases, their applicability to different test species and organ
function interpretation should be evaluated.
6. The value and utility of reversibility studies in a subchronic protocol design needs
to be determined, utilizing data from existing literature for many chemical classes.
7. The evaluation of electroencephalograms in dogs is only a relatively recent art,
and insufficient information is available to allow accurate evaluation. Future
research in this area would serve to rapidly increase the store of knowledge for
this potentially important toxicologic criterion.
8. Methodology for evaluation of audio toxicity should be developed.
9. Methodologies should be developed for evaluating memory and learning and their
interpretation. Because of the increasing number of chemicals that could
potentially affect the higher centers and because of the subtlety of such effects,
techniques to detect early and minor alterations in memory and learning ability
need to be developed and applied on a practical and routine basis.
10. Experimental research is needed to correlate induced sedation and aggressive-
ness with changes in the central nervous system. The changes in the central
nervous system could be either anatomical (destruction of specific areas in the
brain) or changes in the production of chemicals that affect the nerve conduction
and synaptic end points. Little attention has been paid to the central nervous
system effects during the administration of a chemical, and sedation or other
effects have been accepted without consideration of the mechanism involved,
except in the case of pathological lesions.
11. Testing and assessment procedures for cardiovascular toxicity should be
developed, particularly in regard to the causative chemical classes of compounds.
12. The mechanisms of intrahepatic cholestasis and its relationship with hepatic
damage should be studied. This is especially important, because this effect
cannot be predicted accurately by animal studies and because it occurs
frequently in humans.
13. Immunotoxicology methods need to be developed and evaluated for their
potential utility. At the present time, immunology is in the stage of basic research.
Such basic research does not always provide the practical methods necessary for
toxicology. An adequate basic group of immunologic techniques should be
developed that will highlight potential adverse effects on the immune system and
thus trigger the need for further studies.
14. Chemically induced hyperplastic changes in the mammary gland, urinary bladder,
and liver need to be studied to determine their relative significance. Because
these organs, particularly the mammary gland and the liver, respond to various
physiological stimuli, cellular activity varies within the wide normal limits. The
point at which normal hyperplasia becomes abnormal is difficult to determine but
is important relative to evaluating whether a specific change is approaching
preneoplastic or neoplastic transformation. These hyperplastic changes need to
be studied under specific highly controlled conditions in order to develop criteria
that will permit a better evaluation of cellular changes as related to the detection
of carcinogens.
15. The influence of chemical exposure upon spontaneous disease, such as
nonspecific pneumonitis, should be determined. This type of drug effect is
-------
51
important not only in the conduct of toxicology studies, but also as regards the
potential activity of the chemical during exposure of man. Occasionally during
subchronic and chronic toxicity studies, a chemical may appear to affect the
incidence or severity of spontaneous diseases. However, the incidence and
severity of spontaneous disease can vary considerably in animals so that a
determination of chemical effect becomes difficult.
-------
-------
APPENDIX
COMMITTEES AND PARTICIPANTS
Keynote speaker: Warren Muir
Experimental Design
Emil Pfitzer, Chairman
James Allen Bernard McNamara
Domingo Aviado Norbert Page
Kenneth Back Joseph Seifter
George Levinskas Carrol Weil
Relationship of Test Design and
Chemical Classification
Clarence J. Terhaar, Chairman
Sidney Green Peter Voytek
Elliott Harris HanspeterWitschi
Victor Morgenroth III Paul Wright
Marshall Steinberg
Clinical Evaluations and Pathology
Frederick Oehme, Chairman
Diane Beal Donald McGavin
Jerry Kaneko Bobby Joe Payne
Charles Litterst Zophia Zawidzka
Extrapolation from Route to Route
Bob Gibson, Chairman
Werner Braun Arthur McCreesh
Herbert Cornish Daljit Sawhney
Robert Drew
Limitations of Subchronic Studies
Harold Peck, Chairman
James Emerson Ted Loomis
Wayne Galbraith Howard Maibach
Charles Kokoski Carl Smith
53
-------
54
Steering Committee
Wayne Galbraith Emil Pfitzer
Bob Gibson Clarence J. Terhaar
Frederick Oehme Michael Ryon
Norbert Page Daljit Sawhney
Harold Peck
ORNL Support Staff
Norma Callaham Robert Ross
J. Tim Ensminger Michael Ryon
Pat Hartman Donna Stokes
-------
55
Participants
James R. Allen, D.V.M., Ph.D
Department of Pathology
University of Wisconsin
470 North Chaute' Street
Madison, Wisconsin 53706
(608) 263-3524
Domingo Aviado, M.D.
President, Atmospheric Health Sciences,
Inc.
P.O. Box 307
Short Hills, New Jersey 07078
(201) 379-3141
Kenneth C. Back, Ph.D.
Toxicology Branch
Air Force Aerospace Medical Research
Laboratory/ (THT)
Wright-Patterson Air Force Base, Ohio
45433
(FTS) 755-5257
Diane Beal, Ph.D.
Office of Pesticides and Toxic
Substances
Health Review Division—TS/792
Environmental Protection Agency
401 M Street, East Tower
Washington, D.C. 20460
(202) 755-2890
Werner Braun, M.S.
Dow Chemical USA
1803 Building
Midland, Michigan 48640
(517) 636-6151
Herbert H. Cornish, Ph.D.
Department of Environmental and
Industrial Health
University of Michigan
M7525 School of Public Health
Ann Arbor, Michigan 48109
(313) 764-5430
Robert T. Drew, Ph.D.
Medical Department
Brookhaven National Laboratory
Upton, Long Island, New York 11973
(516) 345-3575
James Emerson, D.V.M., Ph.D.*
Manager, Department of Pathology
Abbott Laboratories D469 AP
N. Chicago, Illinois 60064
"Current address:
Manager, Life Sciences
The Coca Cola Company
P.O. Drawer 1734
Atlanta, Georgia 30301
J. Tim Ensminger, M.S.
Health and Environmental Studies
Program
Information Center Complex
Oak Ridge National Laboratory
P.O. Box X, Building 2001
Oak Ridge, Tennessee 37830
(615) 574-7794
Wayne Galbraith, Ph.D.
Office of Research and
Development—RD683
Environmental Protection Agency
401 M Street, SW
Washington, D.C. 20460
(202) 426-2317
James R. Gibson, Ph.D.
E. I. Dupont de Nemours & Co.
Haskell Laboratory for Toxicology and
Industrial Medicine
Newark, Delaware 19711
(302) 366-5257
Sidney Green, Ph.D.
Bureau of Foods
Food and Drug Administration
200 C Street, S.W.
Washington, D.C. 20204
(301) 245-62%
Elliott Harris, Ph.D.
National Institute for Occupational Safety
and Health
4676 Columbia Park
Cincinnati, Ohio 45226
(513) 684-8465
-------
56
Jerry Kaneko, D.V.M., Ph.D.
Department of Clinical Pathology
School of Veterinary Medicine
University of California
Davis, California 95616
(916) 752-0153
Charles J. Kokoski, Ph.D.
Chief of Food Additives Evaluation
Branch
Division of Toxicology
Food and Drug Administration
N. HEW Bldg. 3075
Washington, D.C. 20204
(202) 472-5767
George J. Levinskas, Ph.D.
Director of Environmental Assessment
and Toxicology
Department of Medicine and
Environmental Health
Monsanto Company
800 North Lindbergh Boulevard
St. Louis, Missouri 63166
(314) 694-2184
Charles Litterst, Ph.D.
Laboratory of Toxicology
National Cancer Institute
Bldg. 37, Room 5B-22
Bethesda, Maryland 20205
(301) 496-6713
Ted A. Loomis, Ph.D., M.D.
Department of Pharmacology
University of Washington
School of Medicine
Seattle, Washington 98195
(206) 543-0169
Howard I. Maibach, M.D.
Department of Dermatology
University of California Medical Center
San Francisco, California 94143
(415) 666-2545
(415) 673-9693
Arthur McCreesh, Ph.D.
Toxicology Division
U.S. Arrny Environmental Hygiene
Laboratory
Aberdeen, Maryland 21010
(301) 671-3627
M. Donald McGavin, D.V.M., Ph.D.
Department of Pathobiology
University of Tennessee
College of Veterinary Medicine
P.O. Box 1071
Knoxville, Tennessee 37901
(615) 546-9230
Bernard McNamara, Ph.D.
Chief, Toxicology Branch
Research Division
Chemical Systems Laboratory
Aberdeen Proving Ground, Maryland
21010
(501) 671-3034
Victor H. Morgenroth III, Ph.D.
Food Additives Evaluation Branch
Division of Toxicology
Bureau of Foods
Food and Drug Administration HFF/185
N. HEW Bldg. 3075
Washington, D.C. 20204
(202) 472-5705
Warren R. Muir, Ph.D.
Deputy Assistant Administrator
Office oi Pesticides and Toxic
Substances
Environmental Protection Agency
401 M Street, SW
Washington, D.C. 20460
(202) 755-4894
Frederick Oehme, D.V.M., Ph.D.
Comparative Toxicology Laboratory
College of Veterinary Medicine
Kansas State University
Manhattan, Kansas 66506
(913) 532-5679
-------
57
Norbert P. Page, D.V.M.
Director of Scientific Affairs
Environmental Protection Agency
(TS-792)
401 M Street, SW
Washington, D.C. 20460
(202) 755-4894
Bobby Joe Payne, D.V.M., Ph.D.
Director of Pathology
Toxicity Research Ltd.
510 West Hackley Avenue
Muskegon, Michigan 49444
(616) 733-2584
Harold Peck, M.D.
Safety Assessment Department
Merck, Sharp & Dohme Research
Laboratories
West Point, Pennsylvania 19486
(215) 699-5311
Emil Pfitzer, Ph.D.
Hoffman-LaRoche, Inc.
Experimental Pathology and Toxicology
Nutley, New Jersey 07110
(201)-235-3028
Robert H. Ross, B.S., M.S.
Health and Environmental Studies
Program
Information Center Complex
Oak Ridge National Laboratory
P.O. Box X, Building 2001
Oak Ridge, Tennessee 37830
(615) 574-7797
Michael G. Ryon, B.S.
Health and Environmental Studies
Program
Information Center Complex
Oak Ridge National Laboratory
P.O. Box X, Building 2001
Oak Ridge, Tennessee 37830
(615) 576-2378
Daljit Sawhney, D.V.M., Ph.D.
Environmental Protection Agency
(TS-792)
401 M Street, SW
Washington, D.C. 20460
(202) 755-4864
Joseph Seifter, M.D.
Senior Medical Officer
Office of Testing and Evaluation
Office of Pesticides and Toxic
Substances
Environmental Protection Agency
401 M Street, SW
Washington, D.C. 20460
(202) 755-4894
Carl C. Smith, Ph.D.
Department of Environmental Health
University of Cincinnati
3223 Eden Avenue
Cincinnati, Ohio 45267
(513) 872-5745
Marshall Steinberg, Ph.D.
Vice President and Director of Scientific
Operations
Hazleton Laboratories America, Inc.
9200 Leesburg Turnpike
Vienna, Virginia 22180
(703) 893-5400
Clarence J. Terhaar, Ph.D.
Eastman Kodak Company
Kodak Park—B306
Rochester, New York 14650
(716) 458-1000 ext. 84741
Peter Voytek, Ph.D.
Office of Research and Development
Reproductive Effects Assessment Group,
RD689
Environmental Protection Agency
401 M Street, SW
Washington, D.C. 20460
(202) 426-2275
-------
58
Carrol S. Weil, M.A.
Mellon Institute
4400 Fifth Avenue
Pittsburgh, Pennsylvania 15213
(412) 327-1020
Hanspeter Witschi, M.D.
Biology Division
Oak Ridge National Laboratory
P.O. Box Y, Buildng 9207
Oak Ridge, Tennessee 37830
(615) 574-0801
Paul L. Wright, Ph.D.
Monsanto Company
800 North Lindburg
St. Louis, Missouri 63166
(314) 694-2196
Zophia Zawidzka
Toxicology Research Division
Health Protection Branch
Sir Frederick Banting Building
Health and Welfare Canada
Ottawa, Ontario
Canada K1A OL2
(613) 995-5874
-------
BIBLIOGRAPHY
Adams, R. D., D. Denny-Brown, and C. M. Peterson. 1962. Diseases of Muscle. 2nded. Harper
and Brothers, New York.
Archer, R. K. 1977. Technical Methods. In: Comparative Clinical Haematology. R. K. Archer
and L. B. Jeffcott, eds. Blackwell Scientific Publications, Oxford, pp. 537-610.
Arnold, D. L., S. M. Charbonneau, Z. Z. Zawidzka, and H. C. Grice. 1977. Monitoring Animal
Health During Chronic Toxicity Studies. J. Environ. Pathol. Toxicol. 1:227-239.
Aviado, D. M. 1978. Overwhelming Simularities and Minimal Differences in Toxic Responses of
Rats and Dogs (Part 1:1974-1978 and Part 11:1966-1974). Personal Communication.
Balazs, T. 1970. Measurement of Acute Toxicity. In: Methods in Toxicology. G. E. Paget, ed. F.
A. Davis Company, Philadelphia, pp. 49-81.
Balazs, T., E. Arena, and C. N. Barren. 1972. Protection Against the Cardiotoxic Effect of
Isoproterneol HC1 by Restricted Food Intake in Rats. Toxicol. Appl. Pharmacol.
21:237-243.
Barnes, J. M., and F. A. Denz. 1954. Experimental Methods Used in Determining Chronic
Toxicity. Pharmacol. Rev. 6:191-242.
Bartek, M., J. La Budde, and H. Maibach. 1972. Skin Permeability In Vivo: Comparison in Rat,
Rabbit, Pig, and Man. J. Invest. Dermatol. 58:114-123.
Beckley, J. H., T. J. Russel, and L. F. Rubin. 1969. Use of the Rhesus Monkey for Predicting
Human Response to Eye Irritants. Toxicol. Appl. Pharmacol. 15:1-19.
Bencko, V., and K. Symon. 1969. Dynamics of Arsenic Cumulation in Hairless Mice after
Peroral Administration. J. Hyg. Epidemiol. Microbiol. Immunol. (Prague) 13:248-253.
Bencko, V., and K. Symon. 1970. The Cumulation Dynamics in Some Tissue of Hairless Mice
Inhaling Arsenic. Atmos. Environ. 4:157-161.
Benitz, K. F. 1970. Measurement of Chronic Toxicity. In: Methods of Toxicology. G. E. Paget,
ed. F. A. Davis Company, Philadelphia, pp. 82-131.
Bradley, R. 1978. Skeletal-Muscle Biopsy Techniques in Animals for Histochemical and
Ultrastructrual Examination and Especially for Diagnosis of Myodegeneration in Cattle.
Br. Vet. J. 134:434-444.
Buehler, E. V., and E. A. Newman. 1964. A Comparison of Eye Irritants in Monkeys and
Rabbits. Toxicol. Appl. Pharmacol. 6:701-710.
Cardy, R. H., and J. W. Warner. 1979. Effect of Sequential Bleeding on Body Weight Gain in
Rats. Lab. Anim. Sci. 29(2):179-181.
Cartwright, G. E., 1968. Diagnostic Laboratory Hematology. 4th ed. Grune and Stratton, New
York. 441 pp.
Casarett, L. J. 1975. Toxicologic Evaluation. In: Toxicology: The Basic Science of Poisons. L. J.
Casarett and J. Doull, eds. Macmillan Publishing Company, New York. p. 13.
Case, M. T., J. K. Smith, and R. A. Nelson. 1975. Reproductive Acute and Subacute Toxicity
Studies with Nefopam in Laboratory Animals. Toxicol. Appl. Pharmacol. 33:46-51.
DeLahunta, A. 1977. Veterinary Neuroanatomy and Clinical Neurology. W. B. Saunders Co.,
Philadelphia.
Detweiler, D. K., and D. F. Patterson. 1965. The Prevalence and Types of Cardiovascular
Disease in Dogs. Ann. N.Y. Acad. Sci. 127(1):481-516.
Detweiler, D. K., K. Huben, and D. F. Patterson. 1960. Survey of Cardiovascular Disease in
Dogs—Preliminary Report on the First 1,000 Dogs Screened. Am. J. Vet. Res.
21(82):329-359.
Draize, J. H. 1959. Appraisal of the Safety of Chemicals in Foods, Drugs, and Cosmetics.
Assoc. Food Drug Off. of U.S. 1959:49-51.
Dubowitz, V., and M. H. Brooke. 1973. Muscle Biopsy—A Modern Approach. W. B. Saunders
Company.
59
-------
60
Federal Register. 1978. Proposed Guidelines for Registering Pesticides in the U.S. Hazard
Evaluation: Humans and Domestic Animals. U.S. Environmental Protection Agency [40
CFR Part 163]. 43(163): 37336-37403.
Food Safety Council. 1978. Proposed System for Food Safety Assessment. Columbia, Md. 136
PP-
Gellin, G., M. Ring, and H. Maibach. 1979. Prediction of Human Depigmenting Chemicals with
the Guinea Pig. Contact Dermatitis 5 (In press).
Grant, W. M. 1974. Toxicology of the Eye. Charles C. Thomas, Springfield, 111. 1201 pp.
Harber, L., and A. Shalita. 1977. Immunologically Mediated Contact Photosensitivity in Guinea
Pigs. In: Advances in Modern Toxicology, Vol. 4: Dermatotoxicology and Pharmacology.
F. Marzulli and H. Maibach, eds. Hemisphere Press, Washington, D.C. pp. 427-438.
Hawkey, C. M. 1977. The Haematology of Exotic Mammals. In: Comparative Clinical
Haematology. R. K. Archer, and L. B. Jeffcott, eds. Blackwell Scientific Publications,
Oxford, England, pp. 103-160.
Hoerlein, B. F . 1971. Canine Neurology: Diagnosis and Treatment. 2nd ed. W. B. Saunders Co.,
Philadelphia.
Homan, E. R. 1972. Quantitative Relationship Between Toxic Doses of Antitumor
Chemotherapeutic Agents in Animals and Man. Cancer Chemother. Rep., Part 3.
Janku, I., and M. Krsiak. 1973. Developments and Prospects in Behavioral Toxicology. Proc.
Eur. Soc. Study Drug Toxic. 14:125-129.
Kaneko, J. J., and C. E. Cornelius. 1970. Clinical Biochemistry of Domestic Animals, Vols. I and
II. 2nd ed. Academic Press, New York. Vol. I, 439 pp.; Vol. II, 352 pp.
Kast, A., Y. Tsunenari, M. Honma, J. Nishikawa, T. Shibata, and M. Torii. 1975a. Acute,
Subacute, and Chronic Toxicity Studies of the Beta-Sympathomimetic, Fenterol-HBr on
Rats, Mice, and Rabbits. Oyo Yakuri 10(1):45-71.
Kast, A., Y. Tsunenari, M. Homma, J. Nishikawa, T. Shibata, and M. Torii. 1975b. Acute,
Subacute, and Chronic Toxicity Studies of an Amino-Halogen-Substituted Benzylamine
(Fominoben) in Rats and Mice. Oyo Yakuri 10(l):31-43.
Klassen, C. D. 1975. Absorption, Distribution, and Excretion of Toxicants. In: Toxicology: The
Basic Science of Poisons. L. J . Casarett and J. Doull, eds. Macmillan Publishing Company,
New York. pp. 26-44.
Krasavage, W. J., F. J. Yanno, and C. J. Terhaar. 1973. Dimethyl Terephthalate (DMT): Acute
Toxicity, Subacute Feeding and Inhalation Studies in Male Rats. Am. Ind. Hyg. Assoc. J.
34(l):455-462.
Lewis, S. M. 1967. International Standard for Haemoglobinometry. J. Clin. Path. 20:791-792.
Linman, J. W. 1966. Principles of Hematology. The MacMillan Co., New York. 621 pp.
Litchfield, J. T. , Jr. 1961 . Forecasting Drug Effects in Man From Studies in Laboratory Animals.
J. Am. Med. Assoc. 177:104-108.
Loomis, T. 1974. Essentials of Toxicology. 2nd ed. Lea and Febiger, Philadelphia. 223 pp.
Luna, L. G . , ed. 1968. Manual of Histologic Staining Methods of the Armed Forces Institutes of
Pathology. 3rd ed. McGraw-Hill, New York. 258 pp.
Magnusson, B., and A. Kligman. 1977. Factors Affecting Allergic Contact Sensitization. In:
Advances in Modern Toxicology, Vol. 4: Dermatotoxicology and Pharmacology. F.
Marzulli and H. Maibach, eds. Hemisphere Press, Washington, D.C. pp. 289-301.
Maibach, H., and F. Marzulli. 1977. Phototoxicity (Photoirritation) of Topical and Systemic
Agents. In: Advances in Modem Toxicology, Vol. 4: Dermatotoxicology and Pharmacol-
ogy. F. Marzulli and H. Maibach, eds. Hemisphere Press, Washington, D.C. pp. 211-222.
Marzulli, F., and H. Maibach. 1977. Contact Allergy, Predictive Testing in Humans, In:
Advances in Modern Toxicology, Vol. 4: Dermatotoxicology and Pharmacology. F.
Marzulli and H. Maibach, eds. Hemisphere Press, Washington, D.C. pp. 353-369.
-------
61
Mathias, C. G. T., and H. Maibach. 1978. Dermatotoxicology Monographs. I. Cutaneous
Irritation: Factors Influencing the Response to Irritants. Clin. Toxicol. 13:333-346.
McNamara, B. P. 1976. Concepts in Health Evaluation of Commercial and Industrial
Chemicals. In: Advances in Modern Toxicology, Vol. 1, Part 1: New Concepts in Safety
Evaluation. M. A. Mehlman, R. E. Shapiro, and H. Blumenthal, eds. John Wiley and Sons,
New York. pp. 61-140.
Medway, W., and J. R. Geraci. 1964. Hematology of the Bottlenose Dolphin (Tursiops
Truncatus). Am. J. Physiol. 207:1367-1370.
Meier-Ruge, W. 1977. Workshop on Advances in Experimental Pharmacology of Hydergine,
Basel. 1976. Gerontology 24, Suppl. 1. 153 pp.
National Academy of Sciences. 1975. Principles for Evaluating Chemicals in the Environment.
National Academy of Sciences, Washington, D.C. 454 pp.
National Academy of Sciences. Committee for the Revision of NAS Publication 1138. 1977.
Principles and Procedures for Evaluating the Toxicity of Household Substances. National
Academy of Sciences, Washington, D.C. 130 pp.
Organization for Economic Cooperation and Development. 1979. Final Report of the OECD
Short-Term and Long-Term Toxicology Groups (Unpublished). 185 pp.
Pegg, D. G., J. A. Zempel, W. H. Braun, and P. G. Watanabe. 1979. Dispostion of Carbon-14-
labeled Tetrachloroethylene Following Oral and Inhalation Exposure in Rats. Toxicol.
Appl. Pharmacol. 51(3): 465-474.
Popper, H. P., F. Schaffner, S. Sherlock, and C. M. Levy. 1974. The Liver and Its Diseases.
Intercontinental Medical Books Corporation, New York. 353 pp.
Price, H. M., E. L. Howes, Jr., D. B. Sheldon, O. D. Hutson, R. T. Fitzgerald, J. M. Blumberg,
and C. M. Pearson. 1965. An Improved Biopsy Technique for Light and Electron
Microscopic Studies of Human Skeletal Muscle. Lab. Invest. 14:194-199.
Prieur, D. J., D. M. Young, R. D. Davis, D. A. Cooney, E. R. Homan, R. L. Dixon, and A.M.
Guanno. 1973. Procedures for Preclinical Toxicologic Evaluation of Cancer Chemother-
apeutic Agents: Protocols of the Laboratory of Toxicology. Cancer Chemother. Rep., Part
3, 4(1):1-30.
Reeves, A. L. 1965. The Absorption of Beryllium from the Gastrointestinal Tract. Arch.
Environ. Health 11:209-214
Reeves, A. L., D. Deitch, and A. J. Vorwald. 1967. Beryllium Carcinogenesis. I. Inhalation
Exposure of Rats to Beryllium Sulfate Aerosol. Cancer Res. 27(1)-.439-445.
Robbins, S. L. 1967. Pathology. 3rd ed. Saunders Publishing Co., Philadelphia. 1434 pp.
Smith, C. C. 1950. A Short-Term Chronic Toxicity Study. J. Pharmacol. Exp. Ther.
100:408-420.
Smith, C. C 1979. Extrapolation of Drug Metabolism Data from Animals to Man. In:
Fundamentals and Principles of Clinical Pharmacology. H. P. Kummerle et al., eds. (In
preparation).
Smith, H. A.,T. C.Jones, and R.D. Hunt. 1972. Veterinary Pathology. 4th ed. Lea and Febiger,
Philadelphia, pp. 2-20.
Spurling, N. W. 1977. The Haematology of the Dog. In: Comparative Clinical Haemotology. R.
K. Archer and L. B. Jeffcott, eds. Blackwell Scientific Publications, Oxford, England, pp.
365-440.
Stevens, M. T. 1977. An Alternative Method for the Evaluation of Organ Weight Experiments.
Toxicology 7:275-281.
Stokinger, H. E. and R. L. Woodward. 1958. Toxicologic Methods for Establishing Drinking
Water Standards. J. Am. Works Assoc. 50:515-529.
U.S. Environmental Protection Agency. 1979. Support Document Test Data Development
Standards: Chronic Health Effects. Toxic Substances Control Act, Section 4, EPA 560/11-
79-001, Washington, D.C. 154 pp.
-------
62
Weil, C. S. and D. D. McCollister. 1963. Relationship Between Short- and Long-Term Feeding
Studies in Designing an Effective Toxicity Test. J. Agric. Food Chem. 11(1):486-491.
Weil, C. S., M. D. Woodside, J. R. Bernard, and C. P. Carpenter. 1969. Relationship Between
Single-Peroral, One-Week, and Ninety-Day Rat Feeding Studies. Toxicol. Appl.
Pharmacol. 14:426-431.
Wester, R., and H. Maibach. 1977. Percutaneous Absorption in Man and Animal: A
Perspective. In: Cutaneous Toxicity. V. Drill and P. Lazard, eds. Academic Press, New
York. pp. 111-226.
Wintrobe, M. M., G. R. Lee, D. R. Boggs, T. C. Bithell, J. W. Athens, and J. Foerster. 1974.
Clinical Hematology. 7th ed. Lee and Febiger, Philadelphia. 1896 pp.
World Health Organization. 1978. Acute, Subacute, and Chronic Toxicity Tests. In: Principles
and Methods for Evaluating the Toxicity of Chemicals. Part 1. Environmental Health
Criteria 6. Geneva, pp. 95-115.
Young, J. D., W. H. Braun, and P. J. Gehnng. 1978. Dose-Dependent Fate of 1,4-Dioxane in
Rats. J. Toxicol. Environ. Health 4:709-726.
Zbinden, G. 1963. Experimental and Clinical Aspects of Drug Toxicity. Adv. Pharmacol.
2:1-112.
-------
63
REPORT DOCUMENTATION .' REPORT NO I 2.
PAGE L_EP/L560/lb80_-028 |
4 rule and Subtitle
Proceedings of the Workshop on Subchronic Toxicity
Testing. Denver, Colorado, May 20-24, 1979.
3. Recipient'! Acce»»ion No
s. BwToiu date "oT"
November, 1980 issue
jiawhney, OPTS^ and Michael G. Ry_on_,
Organization Name end AddnKt
Information Center Complex
Information Division
Oak Ridge National Laboratory
Oak Ridge, TN 37830
8, P*rformmf Organization Rept
ORNL/EIS-189
10 Project/Te$k/Work Unit No
11 Contract(C) or Grant(G) No
(c, IAG-80-D-X0453
12. sponsoring Organization Name and Address
Office of Pesticides and Toxic Substances
U.S. Environmental Protection Agency
Uashington, D.C. 20460
13 Type of Report & Pvriod Covcrtd
Final
1C. AMtrict (Lin
This workshop was held at Denver, Colorado, May 20-24, 1979,
to assist the Office of Pesticides and Toxic Substances in developing guide-
lines for Subchronic toxicity testing under the Toxic Substances Control
Act. The participants were organized into committees to discuss the
relationship of protocol design to chemical class, the experimental design
(route of exposure, dose, duration, test species, and acie and nunber of
animals), the evaluation of toxic effects (clinical observations, clinical
laboratory tests, and pathology), the criteria for data extrapolation from
one route of administration to another route, and limitations of acute
and Subchronic tests. Research recommendations Submitted by each committee
for their topic areas are presented in the document.
17. Document Analy*,* • Detcnpton QOS3Q6
Time
Extrapolation
Methodology
Tests
Pathology
Laboratory animals
Subchronic Toxicitv
Health
b Identifier*/Open Ended Tern
c COSATI n*td/Gn>up
EPA Denver Workshop
Subchronic Toxicity Workshop
06/T
It. Availability Statement
Release Unlirmtel
Tl9. Security Cr«» (Thi* Report)
I 20. Security Cl»*t (Tr.it P«f«)
! Unclassified
21 No I
65
(Se« ANSI-239 II)
See instruction* on Reverse
OPTIONAL FORM 272 (4-77)
(Formerly NT1S-35)
Department of Commerce
<*U S GOVERNMENT PRINTING OFFICE 1980 740 062/373
-------
------- |